The Prostate 32:16–26 (1997) Abnormal Prostate Development in C3(1)-bcl-2 Transgenic Mice Xuejun Zhang,1 Min-Wei Chen,1 Albert Ng,1 Po-Ying Ng,1 Chung Lee,3 Mark Rubin,2 Carl A. Olsson,1 and Ralph Buttyan1,2* 1 Department of Urology, College of Physicians and Surgeons, Columbia University, New York, New York 2 Department of Pathology, College of Physicians and Surgeons, Columbia University, New York, New, York 3 Department of Urology, Northwestern University School of Medicine, Chicago, Illinois BACKGROUND. Recent hypotheses to explain the etiology of abnormal growth associated with prostate disease have invoked perturbations in the rate of apoptosis as an important contributor to the onset and progression of these diseases. For this reason, the apoptosis suppressing oncoprotein bcl-2 has come under scrutiny with regards to its role in prostate diseases. In order to evaluate the role of bcl-2 in human prostate disease and to develop an animal model to test anti-bcl-2 therapies, we generated transgenic mice in which bcl-2 expression is targeted to the mouse prostate gland. METHODS. Mouse embryos were microinjected with recombinant DNA constructed by fusing a modified rat C3(1) promotor element to cDNA encoding human bcl-2. Presence of the C3(1)-bcl-2 transgene in progeny was identified by Southern blot and polymerase chain reaction (PCR) analysis. RNase protection assays were used to analyze RNA from 15 organs of these mice. Western blot assays and immunohistochemical staining were used to confirm the tissue-specific protein expression of human bcl-2 and its cellular localization. RESULTS. Three lines of C3(1)-bcl-2 transgenic mice were established. Founder mice carried 2–20 copies of the transgene. Expression of human bcl-2 from the transgene was limited to the prostate gland and testis of males as well as the uterus of females. In the prostate gland, human bcl-2 protein was found only in prostatic epithelial cells. Microscopic analysis of prostate glands from individual males (three lines) showed that these glands were often abnormal, with increased accumulation of cells in the prostatic stroma as well as the epithelium. CONCLUSIONS. These transgenic mice appear to provide a novel animal model for studying neoplastic development of the prostate, with particular emphasis on the bcl-2 protein and the role of apoptosis regulation in such development. Prostate 32:16–26, 1997. © 1997 Wiley-Liss, Inc. KEY WORDS: prostate gland; apoptosis; bcl-2; transgenic mice; prostate-steroid-binding protein INTRODUCTION Benign and malignant growth diseases of the prostate gland rank among the most prevalent afflictions associated with human aging [1,2]. The causative agents involved in the abnormal onset of prostate gland growth in adults are poorly understood. One of the major roadblocks to progress in understanding © 1997 Wiley-Liss, Inc. prostate growth disease has been the lack of animal models for research. Other than humans, dogs are the *Correspondence to: Ralph Buttyan, Department of Urology, College of Physicians and Surgeons, Columbia University, 630 W. 168th Str., New York, NY 10032. Received 7 February 1996; Accepted 18 June 1996 Abnormal Prostates in Transgenic Mice only other animal known to naturally develop prostate hyperplasia with a high frequency . Attempts to establish rodent models of prostatic hyperplasia and malignancy using chemical carcinogenic agents have been reported [4,5], but these animal models are often not suitably reproducible or sufficiently reliable to initiate appropriate studies. More recently, transgenic procedures involving the direct genetic manipulation of mouse embryos with recombinant DNA vectors have produced some lines of mice in which the prostate-directed expression of viral genes has been shown to induce hyperplasia and malignancy [6,7]. We report on the utilization of transgenic technology to produce several independent lines of mice in which the human bcl-2 gene is selectively expressed in the mouse prostate gland. The rationale for the selection of bcl-2 for such experiments comes from a number of recent reports identifying a potential relationship between bcl-2 expression and neoplastic development of the human prostate gland. Immunohistochemical analyses of human prostate tissues showed that bcl-2 protein was not detectable in normal human prostatic secretory epithelial cells, but was highly expressed in prostatic intraepithelial neoplasias (PIN), as well as in a minor subset of untreated prostatic adenocarcinomas and in the vast majority of prostate adenocarcinomas obtained from patients subsequent to hormonedeprivation therapy [8,9]. Furthermore, transfection of the prostate cancer cell line, LNCaP, with a bcl-2 expression vector makes this normally androgensensitive cell line resistant to the effects of androgendeprivation in vitro and when implanted into nude mice . Given this descriptive relationship between increased bcl-2 expression and the development and progression of prostatic cancer to the hormoneresistant state, we wanted to determine the effects of bcl-2 overexpression on prostate gland development in a laboratory animal model. Previous studies showed that the 58 DNA promotor region of the rat ventral prostate secretory protein gene product, C3(1), is suitable for targeting the expression of a chimeric reporter gene (b-galactosidase or SV-40 T-antigen) to the prostate glands of transgenic mice [6,11]. In these earlier experiments, however, the unmodified C3(1) promotor region also permitted promiscuous expression of the reporter gene in other mouse tissues. For this study, we further modified the rat C3(1) promotor region that was utilized for generation of transgenics in the hopes of obtaining a more specific targeting of the mouse prostate gland in transgenic progeny. The three genetic modifications that were made to the C3(1) promotor element included: 1) increasing the length of the 58 promotor region utilized; 2) retention of the first exon and intron 17 of C3(1); and 3) the introduction of site-specific mutations in the two potential translation initiation codons within the first exon. This genetically-modified promotor was fused to the coding region of human bcl-2 cDNA, and this construct was utilized to create several independent lines of transgenic mice. MATERIALS AND METHODS Construction of Chimeric C3(1)-bcl-2 Transgene A 6.4-kb BamHI-PstI restriction fragment containing 4.1 kbp of the upstream 58 promotor region through 17 bp of the second exon of C3(1) were obtained from the C3(1) genomic plasmid p611 (a gift from Dr. M. Parker, Imperial Cancer Research Fund, London, UK) [12,13]. This fragment was subcloned into the pALTER-1 plasmid vector (Promega, Madison, WI) for in situ mutagenesis. Since the first exon for C3(1) contains the authentic ATG translation start codon as well as a second out-of-frame ATG, these two potential ATG codons were mutated to ACGs by the Altered Sites In Vitro Mutagenesis Systemt (Promega). Briefly, this system requires the use of at least two mutagenic oligonucleotides, one to induce a reversion in the mutated ampicillin resistance gene of pALTER, and another to induce a selected mutation within the DNA fragment inserted into pALTER. For our modifications, we needed to place two independent sitespecific mutations in the C3(1) promotor fragment, and this was done by utilizing three mutagenic oligonucleotides, one provided with the kit (to revert the ampr gene of pALTER), and two additional ones (synthesized by National Biosciences, Inc., Plymouth, MN) to insert mutations at the sequential ATG sites in the first exon of C3(1), i.e., 58-GCCTCAACACGAAGCTGG-38 and 58-TGCTGCTACGCCAGTGGTAA-38. These oligonucleotides were simultaneously annealed to the single-stranded DNA template of the pALTER vector. The annealed DNA was treated with T4 DNA polymerase to extend the primers and was then ligated with T4 DNA ligase. The treated vector was used to transform competent BMH 71-18 cells, and bacteria-containing mutated vectors were selected by growth on ampicillin agar plates. Resistant colonies were picked and the plasmids were sequenced by use of standard dideoxynucleotide methods. A colony that contained both mutations within the first exon was selected for further work. The modified C3(1) promoter from this colony was subcloned into BamHI-PstI digested pBluescript SK vector (Stratagene, La Jolla, CA). The DNA insert from the pSFFV/bcl-2 plasmid (obtained from Dr. S. Korsmeyer, Washington University School of Medicine, St. Louis, MO)  containing the 1.8-kb human bcl-2 cDNA, 0.9-kb SV-40 early 18 Zhang et al. splice, and SV-40 polyadenylylation signal was excised with EcoRI (partial digestion)-HindIII and subcloned into the downstream EcoRI-HindIII (partial digestion)-digested C3(1)-pBluescript vector. This construct was called C3(1)-bcl-2, and the correct transcriptional orientation of the bcl-2 reporter gene was confirmed by DNA sequencing by use of standard dideoxynucleotide methods. Experimental Animals and Production of Transgenic Mice Animals used in these experiments were maintained within the Institute of Comparative Medicine at the Columbia University Health Sciences Division. All experiments were conducted in accordance with the highest standards of humane animal care, as outlined in the NIH Guide for the Care and Use of Laboratory Animals. Transgenic mice were generated as described . The entire C3(1)-bcl-2 transgene was excised from the vector with NotI and ClaI. The fragment was isolated by gel electrophoresis followed by electroelution from the gel, and was further purified by CsCl ultracentrifugation. Purified DNA was resuspended in microinjection buffer (10 mM Tris-HCl, pH 8.0, 0.1 mM EDTA) to a final concentration of 2–5 ng/ ml and used for microinjection into the male pronucleus of fertilized mouse oocytes derived from the C57BL/6J × CBA/J mouse strain (Jackson Laboratory, Bar Harbor, ME). Microinjected embryos were maintained in culture for 24 hr prior to implantation into pseudopregnant female recipients. Litters were born to these mice 21 days later, and the progeny were maintained with the surrogate mothers until weaning. Portions of the tail were clipped at this time for extraction of DNA. Identification of Transgenic Mice Transgenic progeny were identified by Southern blot analysis of tail DNA isolated from 3-week-old litters using standard techniques . Ten micrograms of tail DNA were digested with EcoRI and electrophoresed through a 1% agarose gel. The DNA was transferred to nylon membranes (Boehringer Mannheim, Inc., Indianapolis, IN) and hybridized with the 1.8-kb human bcl-2 cDNA probe labeled with 32P-dCTP, using the Random Primed DNA Labeling Kitt (Boehringer Mannheim, Inc.). The membranes were washed under high stringency conditions, and were exposed to X-ray film at −70°C overnight. Transgenic progeny were identified by the presence of a 1.8-kb human bcl-2 cDNA band, and the number of copies was calculated by comparing the film band density obtained from the mouse endogenous bcl-2 gene with the intensity of the human bcl-2 band performed with the Molecular Dynamics Scanning Laser Densitometer (Molecular Dynamics, Sunnyvale, CA). A PCR-based technique was then used to confirm positive offspring transgenic mice, as previously described . The primers used in PCR screening were designed to fall within the C3(1) first intron (A) and human bcl-2 cDNA (B) (detailed in Fig. 1). The sequences of the synthetic oligonucleotides were as follows: A, 58GCCCATCACCTTGCTTAT-38; B, 58-CACATCTCCCGCATCCCACT-38. These primers produce a 283-bp DNA fragment following PCR amplification of the recombinant C3(1)-bcl-2 transgene. RNA Isolation and Analysis Total cellular RNA was isolated from transgenic mouse organs including brain, heart, kidney, spleen, lung, thymus, salivary, bladder, gut, prostate, seminal vesicles, testis, muscle, uterus, and mammary gland, using the RNzole B reagent (TelTest, Inc., Friendswood, TX), following the manufacturer’s instructions. RNase protection assays were performed on these RNAs as previously described , with the RPA II Ribonuclease Protection Assay Kitt (Ambion, Inc., Austin, TX). A small cDNA fragment of 283 nucleotides, encoding a portion of the C3-bcl-2 transgene, was obtained by PCR amplification from the intact C3(1)-bcl-2 plasmid, as mentioned above. The amplified DNA fragment was cloned into the TA II cloning vector (Invitrogen, San Diego, CA), and was sequenced using standard dideoxynucleotide methods. This fragment will protect a chimeric mRNA sequence containing the 17 nucleotides from C3(1) exon 2 and the adjacent 163 nucleotides from human bcl-2 cDNA (see Fig. 1). Synthesis and purification of the 32Plabeled antisense riboprobe from this vector were done as previously described . Twenty micrograms of total RNA were denatured and hybridized overnight at 45°C to labeled antisense RNA transcribed in vitro in the presence of 32P-UTP. RNase treatment was performed at 37°C for 1 hr. Digests were applied to 8% acrylamide sequencing gels and electrophoresed at 350 V for 3 hr. The gel was exposed overnight to X-ray film to produce an autoradiograph. Western Blot Analysis Transgenic and normal control mouse prostate, testis, and uterus removed from necropsy were frozen in liquid nitrogen, powdered, and homogenized in lysis buffer (50 mM Tris-HCl, pH 8.0, 150 mM sodium chloride, 0.1% sodium dodecyl sulfate, 1% NP-40, and 0.5% sodium deoxycholate) on ice. Proteinase inhibitors, including PMSF (100 mg/ml), pepstatin A (1 mg/ ml), aprotinin (1 mg/ml), and leupeptin (1 mg/ml) Fig. 1. Top: Diagrammatic description of DNA elements involved in construction of the recombinant C3(1)-bcl-2 gene. A genetically modified (mutated) fragment containing 4.1 kbp DNA upstream from the transcriptional start site of C3(1), as well as the first exon, first intron, and a small portion (17 bp) of the second exon, were fused to a 1.85-kbp cDNA fragment of human bcl-2  and the early splice site and poly A signal site for SV-40. Bottom: Partial sequence of exon 1 of C3(1), showing two sites where potential translation start codons (ATGs) were mutated to ACGs by an in situ mutagenesis system. a and b (top) represent approximate sites for binding of PCR primers, as described in Materials and Methods. 20 Zhang et al. (Sigma Chemical Co., St. Louis, MO) were added to the lysis buffer prior to homogenization. Insoluble debris was removed by centrifugation at 10,000g for 10 min at 4°C. Protein concentrations were determined in these extracts using the Bio-Rad Protein Assay System (Bio-Rad Labs, Inc., Richmond, CA). Aliquots of cell extracts containing 50 mg of protein were electrophoresed on an 8% Laemmli SDS-polyacrylamide gel at 150 V for 45 min and electrophoretically transferred to a nitrocellulose filter (Amersham Life Science, Arlington Heights, IL) at 110 V for 60 min in 25 mM TrisHCl, pH 8.0, 0.192 M glycine, and 20% methanol. The filter was blocked in TBS-T buffer (20 mM Tris-HCl, pH 8.0, 0.136 M NaCl, 5% nonfat milk, and 0.5% Tween-20) at 4°C for 1 hr and then incubated overnight with a mouse monoclonal anti-human-bcl-2 antibody (Dako bcl-2, 124; Dako Corp., Carpinteria, CA)  diluted to 1:2,000 in TBS-T. After a series of washes in TBS-T buffer, the filter was incubated with the secondary antibody (sheep anti-mousehorseradish peroxidase complex), supplied by Amersham Life Science, Inc. in the ECL Western Blotting Analysis System (Amersham Life Science, Inc.). Chemiluminescent detection of antibody was accomplished using the reagents provided in this system, following exposure of the blot to Kodak XAR-5 X-ray film (Eastman Kodak, Rochester, NY) for 15 sec. Immunohistochemical Studies of Transgenic Mice Tissues of prostate, testis, and uterus from nontransgenic control and transgenic mice were collected at necropsy and fixed in 10% neutral buffered formalin. The tissue was dehydrated and embedded in paraffin, and 5-m thin sections were cut from these tissues with a microtome. The sections were deparaffinized with xylene and rehydrated in a graded series of ethanol solutions. The immunostaining analysis for transgenic human bcl-2 protein was performed with the HistoMouse Kit (Zymed Laboratories, Inc., South San Francisco, CA), which was designed to detect reactivity of mouse primary antibodies on rodent tissue without background. Briefly, tissue sections were treated with 3% peroxidase quenching solution for 10 min. Nonspecific background was eliminated by incubating the slides with blocking solution A and blocking solution B. The mouse monoclonal anti-human bcl-2 antibody (Dako 124, Dako Corp.) was incubated on the tissues, followed by addition of a biotinylated secondary antibody (reagent 1C). Streptavidinperoxidase was then added (reagent 2) to bind to the biotin residue on the linking antibody. The presence of peroxidase was revealed by addition of substratechromogen solution (reagents 3A–C). Then peroxide was utilized to convert the substrate to a red insoluble deposit, which demonstrates the location of the transgenic human bcl-2 protein. Histology Tissues for light microscopy were collected from sacrificed animals. Some tissues were fixed and embedded, as described above for preparation of thin sections, while other tissues were embedded in OCT compound and rapidly frozen to −70°C in a biopsy mold. Thin sections (8 m) were obtained from the frozen tissues by a cryostat. All sections were stained with hematoxylin and eosin for microscopic analysis. RESULTS Generation of C3-bcl-2 Transgenic Mice C3(1) is a rat gene encoding the C3 peptide subunit of the major secretory protein of the ventral prostate gland, PSP . Previously, we reported some success in targeting b-galactosidase (b-gal) expression to the prostate glands of transgenic mice with the use of a chimeric DNA molecule that was constructed by fusing a portion of the 58 promotor region of the rat C3(1) gene to the coding region for bacterial b-gal . In some founder lines made with this DNA construct, the C3(1) promotor also allowed promiscuous expression of the b-gal reporter in the seminal vesicles as well as in the testis. Likewise, another report describing the use of a C3(1) promotor region to target SV-40 Tantigen expression to the prostate gland also identified promiscuous expression of the transgene product in thyroid, salivary gland, and cartilage, as well as in the breast tissue of females . In an attempt to further restrict the tissue-specific expression of a reporter gene, we first made genetic modifications to the C3(1) promotor element (detailed in Fig. 1) that was subsequently utilized to target the expression of a human bcl-2 reporter gene to the mouse prostate gland. The modifications that we made increased the amount of 58 upstream (C3(1) promotor) DNA and ensured the retention of any potential genetic regulatory element within the first intron of rat C3(1), an area containing a putative site for the presence of an androgenreceptor binding and response element . The promoter sequences of C3(1) that were utilized in our experiments are contained within a DNA fragment generated by BamHI and PstI digestion of the C3(1) genomic clone, 611 [12,13]. This fragment contains 4.1 kbp of the 58-flanking C3(1) promoter sequence, as well as the first exon, entire first intron, and a small 58 fragment of exon 2. The two potential ATG translation initiation sites within the first exon were sequentially mutated (to ACG) to prevent inappropriate translation Abnormal Prostates in Transgenic Mice 21 specimens examined. For mice in which the transgene was integrated into genomic DNA, we also expected to find hybridization to a fragment of 1.85 kbp. This fragment was detected in digested tail DNA extracted from progeny animals 3, 5, 15, 19, and 21 (Fig. 3A, lanes 3–7). No hybridization to a 1.85-kb fragment was detected in the DNA extracted from a control mouse (Fig. 3). Additional and distinct hybridizing bands were found in DNAs obtained from mice 3, 5, 19, and 21, demonstrating that the transgenic bcl-2 was inserted at separate sites of the mouse genome in these lines. A comparison of the transgenic bcl-2 band intensities to the intensity of the normal mouse genomic bcl-2 gene suggests that about 2–20 copies were incorporated in the various transgenic progeny lines. A PCR amplification technique was also used to screen DNA from the putative positive transgenic mice. All 5 transgenic founder mice demonstrated the expected 283-bp transgene PCR product (Fig. 3B). Tissue-Specific Expression of Human bcl-2 in Transgenic Mice Fig. 2. Autoradiograph of DNA sequencing gel demonstrates introduction of mutations in the two ATG sites within the first exon of C3(1) prior to construction of the C3(1)-bcl-2 transgene. In this manipulation, ATG sites were mutated to ACGs. of the recombinant transgene message by means of in situ mutagenesis. DNA sequencing was used to confirm that both the authentic ATG translation start codon and the out-of-frame ATG sites were mutated to ACGs in the final modified promotor fragment (Fig. 2). The transgene of C3-bcl-2 was completed by enzymatic ligation of DNA fragments, including the modified 58 C3(1) promoter, human bcl-2 cDNA, and the 38 SV-40 polyadenylation signal (Fig. 1). The hybrid C3(1)-bcl-2 gene fragment was purified by CsCl ultracentrifugation and was microinjected into the pronuclei of C57BL/6 × C57BL/F1 fertilized eggs. The twocell-stage embryos were reimplanted into pseudopregnant C57BL/6 outbred females, and the females were maintained through the birth and subsequent weaning of the pups. Analysis of DNA obtained from tail biopsies of the pups identified 5 potential founder mice by Southern blot analysis. Tail DNAs were digested with EcoRI restriction endonuclease, electrophoresed, and blotted onto nitrocellulose paper. The blot was hybridized to a radiolabeled 1.85-kbp human bcl-2 cDNA probe (Fig. 3A). Two EcoRI bands (5.7 kb and ∼10 kb), corresponding to hybridization with the endogenous mouse bcl-2 gene fragments, were seen in all DNA To test the potential of the modified C3(1) promotor fragment for targeting prostate gland-specific expression of bcl-2, we analyzed the RNA of numerous tissues obtained from first-generation (heterozygous) transgenic offspring for expression of hybrid C3-bcl-2 sequences by an RNase protection assay. An antisense riboprobe that would protect transgene mRNA containing 17 nucleotides of C3(1) exon 2 fused to 163 nucleotides of human bcl-2 cDNA was used to detect the presence of this messenger RNA in various organs (see Fig. 4A). This riboprobe is expected to protect (following hybridization) a 180-nucleotide fragment from RNase digestion when the transgene is expressed. The 180-nucleotide-protected bands were detected when test RNAs were obtained from the prostate glands and testis of male progeny mice as well as from the uterus of female progeny mice in three different transgenic mouse lines (3, 5, and 21). An example of tissues (male and female) analyzed from mice in the second generation of line 5 is shown in the autoradiograph of Figure 4B. No protected fragments were obtained when RNAs were obtained from tissues (including prostate, uterus, and testis) of nontransgenic littermate mice (Fig. 4B). Expression of human bcl-2 protein was confirmed in these same tissues of three transgenic lines by Western blot analyses. In Figure 5, a Western blot probed with a monoclonal antibody against human bcl-2 protein revealed the presence of the 26-kd bcl-2 protein in an extract of the derivative LNCaP cell line (LNCaP-bcl-3) that overexpresses bcl-2 subsequent to transformation , as well as in protein extracts obtained from the prostate 22 Zhang et al. Fig. 3. A: Autoradiograph of Southern blot demonstrates hybridization of bcl-2 cDNA to a 1.85-kbp DNA fragment present in transgenic mouse founder lines, as well as to the endogenous mouse bcl-2 gene fragments at 10 kbp and 5.2 kbp. B: Ethidium bromide staining pattern of an agarose gel following electrophoresis of reaction products obtained from a PCR amplification reaction using DNA obtained from a control (nontransgenic) mouse tail (lane 1), or from DNAs extracted from the tails of founder mouse lines, as indicated. A DNA fragment at 283 bp was generated from amplification of the chimeric C3(1)-bcl-2 gene. glands, testis, and uterus of transgenic mice (lines 3, 5, and 21). The bcl-2 protein was abundantly expressed in these tissues but was absent from the comparable tissues obtained from nontransgenic mice (Fig. 5). Immunohistochemical Analysis of Transgenic Mouse Tissues A mouse monoclonal antibody against human bcl-2 was used to immunostain prostate, testis, and uterus from heterozygote transgenic progeny (lines 3, 5, and 21). This antibody distinctly stained epithelial cells of the ventral prostate (Fig. 6), as well as interstitial (fibroblastic) cells of the testis and uterus (not shown). Equivalent application of this staining protocol to sections of these same tissues obtained from control (nontransgenic) littermates did not identify any immunostaining cells in these tissues. Abnormalities of Prostate Gland Detected in Transgenic Progeny Mice Overall, the size and weights of prostate glands (as well as the other male urogenital tract organs) of heterozygote transgenic male mice were virtually indistinguishable from tissues obtained from nontrans- genic, age-matched males. Nor was there any evidence for increase in the proliferative index of these tissues, based on overt counting of mitotic cells in thin sections of tissues. Nonetheless, the prostate glands obtained from some of these mice showed a peculiar morphological appearance when examined by microscopy. As shown in Figure 7, a ventral prostate gland isolated from one particular transgenic male mouse (line 5, 3 months old) was distinctly affected, having an extensive accumulation of cells both within the stromal compartment as well as in the epithelial cell compartment. In spite of the fact that the prostate gland was not significantly enlarged compared to normal mouse prostate gland, microscopic analysis showed that the cellular elements were crowded to the extent that very little luminal space was evident in the prostatic ducts. Ventral prostate tissues from 6 of 14 other individual males of this line were also affected to a lesser degree, as were 3 of 9 males analyzed from line 21, and 1 of 5 males analyzed from line 3. The frequency with which this phenotype is observed in males of the different lines (line 5 > line 21 > line 3) reflects the relative ranking of these lines for transgene copy numbers (∼20 copies, line 5; ∼4 copies, line 21; and ∼2 copies, line 3). In summary, prostate tissues obtained from individual transgenic males from any Abnormal Prostates in Transgenic Mice 23 Fig. 4. A: Diagram identifies potential fragment of C3-bcl-2 transgene RNA that will be protected by antisense riboprobe generated by in vitro transcription of a DNA fragment produced by PCR, utilizing primers a and b. B: RNase protection assay identifies expression of the chimeric C3(1)-bcl-2 gene in tissues obtained from 4-month-old transgenic mice from first-generation heterozygotes of line 5. The 180-bp protected fragment was observed in RNA extracted from the prostate, testis, and uterus of this transgenic line, but not in RNA extracted from the same tissues of nontransgenic (control) mice. given line showed variation in this characteristic. Almost half the males from line 5 were affected, whereas a decreasing proportion of males from lines 21 and 3 (having fewer copies of the transgene) showed signs of this atypical prostatic hypertrophy. For mice that were affected, prostatic hypertrophy involved both stromal and epithelial elements, and this is enigmatic considering that the transgene product (human bcl-2) is only expressed in the epithelial cells. The two other tissues that consistently demonstrated expression of human bcl-2, testis and uterus, had no discernible phenotypic change detectable by size or microscopic analysis in any of the male or female progeny. DISCUSSION The bcl-2 protein is a potent suppressor of apoptosis [21,22]. This protein is well-known for its role in the development of certain forms of human lymphomas . Likewise, it is suspected of having a natural role in the development and progression of cancers of the human breast and prostate gland [8,9,24]. The transgenic mice which we produced in these experiments, having greatly elevated amounts of human bcl-2 protein expressed in the prostate gland, should help in identifying any potential neoplastic or malignant phenotype associated with bcl-2 overexpression in the prostate. In our experiments, targeting of bcl-2 protein expression to the mouse prostate gland in transgenic animals was accomplished by use of a geneticallymodified promotor element from the rat C3(1) gene. These modifications included an increase in the amount of C3(1) DNA upstream from the 58 transcription start site utilized for the recombinant transgene. Previously, we had used a smaller promotor element 24 Zhang et al. Fig. 5. Immunochemical analysis of a Western blot identifies the 25-kd human bcl-2 protein in extracts of prostate tissue, testis, and uterus obtained from first-generation heterozygotes of line 5, but not from comparable tissue obtained from nontransgenic mice. derived from this gene to target the expression of a neutral reporter gene (b-gal) and had found that the 58 C3(1) promotor allowed for promiscuous expression of the reporter gene in other male tissues (testis and seminal vesicle) . Likewise, in another study in which the C3(1) promotor was utilized to target the expression of SV-40 T-antigen to the prostate, promiscuous expression of the T-antigen was found in a number of other transgenic mouse tissues . Secondly, because genetic regulatory elements (an androgen receptor binding site) were also described in the first intron of the C3(1) gene , we had hoped that genetic manipulations of this 58 C3(1) promotor, so that the first intron would be maintained, might increase the specificity of targeting reporter genes to the prostate gland in transgenic mice. Since the first intron contains two ATG codons (potential translation start sites), these sites were mutated to ACGs so that translation of the reporter gene must be initiated by an appropriate codon within the reporter cDNA. In the three independent lines of transgenic mice that we have so far characterized, the modified C3(1) promotor element containing the first exon and intron still allows for promiscuous expression of the bcl-2 reporter gene (in our lines, to the interstitial cells of the testis and uterus). The reason for this promiscuous expression is not clear, since C3(1) mRNA or protein is not normally expressed in mouse testis or uterus . The promiscuous expression of the bcl-2 reporter protein, however, does not seem to overtly affect the phenotype of the testis or uterus of the transgenic mouse lines in the same way as the prostate gland. Perhaps this is because testicular and uterine fibroblastic cells have such a normally long life span (and low cell turnover) that the increased expression of bcl-2 here does not further affect the tissue. To date, a minor proportion (20–50%) of male heterozygous progeny mice from at least three different (independent) lines of C3(1)-bcl-2 transgenic mice has shown benign morphological changes in the prostate glands, involving increased cellular content. This phenotype is variable, appearing more frequently in lines with a higher transgene copy number, and with differing severity in individual mice. The affect was most pronounced in males from line 5, and it appears that the increased number of human bcl-2 copies carried by this transgenic line influences both the frequency of prostatic hypertrophy and the severity of this phenotype. Since human bcl-2 expression was found to be limited to the epithelial cells in these mice, this phenotype is enigmatic but extremely interesting, because it involves increased stromal as well as epithelial cell content of the gland. The protein encoded by the bcl-2 gene is known to suppress apoptosis, and it is reasonable to speculate that the overexpression of bcl-2 in the epithelial cells of transgenic mouse prostate glands is increasing their life span, resulting in abnormal accumulation of epithelial cells as the mouse ages. However, we have not found any evidence that the human bcl-2 gene is expressed in the stromal cells of these transgenic mouse prostate glands, and so the cause of the accumulation of stromal cells in these same tissues is likely not due to the direct action of bcl-2 protein. In the future, examination of these mice might provide evidence for the production of excess stromal cell growth factors by the aberrantly long-lived prostatic epithelium . Likewise, this development of a benign hypertrophic condition in the prostate glands of some of these transgenic mice provides the basis for further analysis as to whether bcl-2 might be an important factor in the development of human BPH. A preliminary study Abnormal Prostates in Transgenic Mice 25 Fig. 6. Immunohistochemical staining of ventral prostate tissue obtained from a first-generation heterozygote male from line 5 shows localization of the human bcl-2 protein to the epithelial cells. The anti-bcl-2 antibody used in this immunostaining protocol does not stain sections of control (nontransgenic) mouse prostate glands. ×200. Fig. 7. A: Microscopic analysis of thin sections made from ventral prostate glands obtained from control mouse prostate gland. B: Severely affected male mouse (3 months old) from line 5 shows phenotypic evidence for cellular accumulation in this specimen. ×100. already suggests that bcl-2 expression might be elevated in epithelial cells of human BPH tissue . In summary, these transgenic mice provide a novel animal model for studying neoplastic development of the prostate, with particular emphasis on the role of apoptosis regulation in such development. At this time, all the progeny mice that we have analyzed have been heterozygotes. It is possible that homozygotic progeny will have increased production of human bcl-2 protein in the prostate, and therefore will develop a more pronounced and consistent abnormal prostate phenotype. Likewise, these animals will be 26 Zhang et al. followed through their aging process to determine whether increased bcl-2 production might be a factor in the subsequent development of prostate cancers. ACKNOWLEDGMENTS This work was supported by grants from the National Institutes of Health (CA48089), the CaPCure Foundation, the David Koch Foundation, the New York Academy of Medicine, and the ColumbiaPresbyterian Medical Center Urological Research Fund. REFERENCES 1. McNeal J: Pathology of benign prostatic hyperplasia: Insight into etiology. Urol Clin North Am 17:477–496, 1990. 2. Catalona WJ: Management of cancer of the prostate. N Engl J Med 232:335–336, 1995. 3. Coffey DS, Walsh PC: Clinical and experimental studies of benign prostatic hyperplasia. Urol Clin North Am 17:461–475, 1990. 4. Pollard M, Luckert PH, Snyder DL: The promotional effect of testosterone on induction of prostate cancer in MNU-sensitized L-W rats. Cancer Lett 45:209–212, 1989. 5. Bosland MC, Prinsen MK: Induction of dorsolateral prostate adenocarcinomas and other accessory sex gland lesions in male Wistar rats by a single administration of N-methyl-Nnitrosourea, 7,12-dimethylbenz (a) anthracene, and 3,28dimethyl-4-aminobiphenyl after sequential treatment with cyproterone acetate and testosterone propionate. Cancer Res 50: 691–699, 1990. 6. Maroulakou I, Anver M, Garrett L, Green JE: Prostate and mammary adenocarcinoma in transgenic mice carrying a rat C3(1) simian virus 40 large tumor antigen fusion gene. Proc Natl Acad Sci USA 194:11236–11240, 1994. 7. Greenberg NM, DeMayo F, Finegold MJ, Medina D, Tilley WD, Aspinall JO, Cunha GR, Donjacour AA, Matusik RJ, Rosen JM: Prostate cancer in a transgenic mouse. Proc Natl Acad Sci USA 92:3439–3443, 1995. 8. Colomel M, Symmans F, Gil S, O’Toole KM, Chopin D, Benson M, Olsson CA, Korsmeyer S, Buttyan R: Detection of apoptosissuppressing oncoprotein bcl-2 in hormone-refractory human prostate cancers. Am J Pathol 143:390–400, 1993. 9. McDonnell TJ, Troncoso P, Brisbay SM, Logothetis C, Chung LW, Hsieh JT, Tu SM, Campbell ML: Expression of the protooncogene bcl-2 in the prostate and its association with emergence of androgen-independent prostate cancer. Cancer Res 52:6940– 6944, 1992. 10. Raffo AJ, Perlman H, Chen MW, Day ML, Streitman JS, Buttyan R: Overexpression of bcl-2 protects prostate cancer cells from apoptosis in vitro and confers resistance to androgen depletion in vivo. Cancer Res 55:4438–4445, 1995. 11. Buttyan R, Slawin K: Rodent models for targeted oncogenesis of the prostate gland. Cancer Metastasis Rev 12:11–19, 1993. 12. Hurst HC, Parker MG: Rat prostatic steroid binding protein: DNA sequence and transcript maps of the two C3 genes. EMBO J 2:769–774, 1983. 13. Parker MG, White R, Hurst H, Needham M, Tilly R: Prostatic steroid-bind protein: Isolation and characterization of C3 genes. J Biol Chem 258:12–15, 1983. 14. Hockenberry DM, Nunez G, Milliman C, Schreiber RD, Korsmeyer SJ: Bcl-2 is an inner mitochondrial membrane protein that blocks programmed cell death. Nature 348:334–336, 1990. 15. Hogan B, Costantini F, Lacy E: ‘‘Manipulating the Mouse Embryo.’’ Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, 1986. 16. Sambrook JE, Fritsch J, Maniatis T: ‘‘Molecular Cloning: A Laboratory Manual.’’ Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, 1989. 17. Chen S, Even GA: A simple screening method for transgenic mice using the polymerase chain reaction. Biotechniques 8:32– 34, 1990. 18. Zhang X, Colombel M, Raffo A, Buttyan R: Enhanced expression of p53 mRNA and protein in the regressing rat ventral prostate gland. Biochem Biophys Res Commun 198:1189–1194, 1994. 19. Heyns W, Peeters B, Mous J, Bossyns D, Rombauts W, de Moor P: Prostatic binding protein and its hormonal regulation. Prog Clin Biol Res 75:339–350, 1981. 20. Tan JA, Marschke KB, Ho KC, Perry ST, Wilson EM, French FS: Response elements of the androgen-regulated C3 gene. J Biol Chem 267:4456–4466, 1992. 21. Korsmeyer SJ, Shutter JR, Veis DJ, Merry DE, Oltvai ZN: Bcl-2/ Bax: A rheostat that regulates an anti-oxidant pathway and cell death. Semin Can Biol 4:327–332, 1993. 22. Reed JC: Bcl-2 and the regulation of programmed cell death. J Cell Biol 124:1–6, 1994. 23. Tsujimoto Y, Cossman J, Jaffe E, Croce CM: Involvement of the bcl-2 gene in human follicular lymphoma. Science 228:1440– 1443, 1985. 24. Leek RD, Kaklamanis L, Pezzella F, Gatter KC, Harris AL: Bcl-2 in normal human breast and carcinoma: Association with oestrogen receptor-positive, epidermal growth factor receptornegative tumors and in situ cancer. Br J Cancer 691:135–139, 1994. 25. Chung LW: The role of stromal-epithelial interaction in normal and malignant growth. Cancer Surv 23:33–42, 1995. 26. Berchem G, Cardillo M, Tarkington MA, Vicker C, Tehan TJ, Ortega LG, Saini N, Patterson RH, Lage J, Gelman EP: Bcl-2 and apoptosis in benign and malignant human prostate neoplasia after acute androgen withdrawal. Cancer Res 37:243, 1996.