The Prostate 2920-29 ( I 996) Parathyroid Hormone-Related Protein (PTHrP) is an Epidermal Growth Factor-Regulated Secretory Product of Human Prostatic Epithelial Cells Scott D. Cramer, Donna M. Peehl, Michelle G. Edgar, Stephen T. Wong, Leonard J. Deftos, and David Feldman Departments of Urology (S.D.C., D.M. P., M.G.€, S.T.W.) and Medicine (D.F.), Stanford University School of Medicine, Stanford, California; and Department of Medicine, University of California and the San Diego VA Medical Center, Son Diego California (Lj.D.) Parathyroid hormone-related protein (PTHrP)has previously been shown to be expressed in human prostatic tissue and in prostatic cancer cell lines. In the present study, PTHrP immunoreactivitywas detected in the glandular epithelium of normal prostate and benign prostatic hyperplasia (BPH), as well as in prostatic adenocarcinoma (Cap). Epithelial cell cultures derived from normal, BPH, and CaP tissues were also stained by antibodies against PTHrP, and northern analysis revealed multiple transcripts of PTHrP in the cellular RNA. PTHrP (1-34) was measurable by radioimmunoassay (RIA) in media conditioned by the prostatic epithelial cell cultures, and PTHrP accumulated in conditioned media during a 72 hr time course. Addition of complete growth medium to starved cells resulted in increased PTHrP mRNA levels by 1 hr, with maximal stimulation at 8-24 hr. Several individual factors contained in the complete growth medium were tested for their ability to regulate PTHrP expression. Epidermal growth factor (EGF) was the major inducer of PTHrP expression, while cholera toxin, bovine pituitary extract, hydrocortisone, and insulin had minimal or no effect on PTHrP transcript levels. Since each of these factors is growth stimulatory, the unique ability of EGF to induce PTHrP is apparently unrelated to mitogenicity. 1,25-DihydroxyvitaminD, [1,25(OH),D,], an inhibitor of PTHrP expression in several other cell types, had no effect on steady-state levels of PTHrP mRNA expressed by epithelial cells in complete growth medium, although prostate cells have vitamin D receptors and are responsive to 1,25(OH),D, in other ways. Our results indicate that PTHrP expression is not confined to the neuroendocrine cells of the human prostate and that our culture system can be used as a model to investigate the role of PTHrP in the prostate. 0 1996 Wiley-Liss, Inc. ABSTRACT: KEY WORDS: 0 1996 Wiley-Us, Inc. immunohistochemistry, radioimmunoassay, Northern analysis, 1,25-dihydroxyvitamin D, PTHrP in Human Prostatic Cells tic . Identification of factors produced by the prostate that are known to have direct actions on bone remodeling could be important for understanding the nature of the processes involved in prostatel bone interactions. Parathyroid hormone-related protein (PTHrP) is generally considered to be the primary factor responsible for humoral hypercalcemia of malignancy . The hypercalcemic actions of PTHrP occur via stimulation of renal distal tubular calcium reabsorption and increased osteoclastic bone resorption . These effects of PTHrP are thought to be mediated through a common PTWpTHrP receptor [q.In bone, M w PTHrP receptor mRNA has been demonstrated in 0steoblasts and chondrocytes  but not in osteoclasts, suggesting an indirect effect of PTHrP on osteoclastic bone resorption. PTHrl' expression in normal bone is thought to be important for the development of normal bone architecture . PTHrP gene expression in breast cancer has been shown to be higher in tumors that eventually metastasize to bone as opposed to breast tumors that metastasize to other nonbone sites [lo]. Overexpression of PTHrP gene expression in breast cancer cell lines can enhance osteolytic bone metastases [ll].These studies suggest that PTHrP expression might be linked to breast cancer metastasis to bone. Recent identification of PTHrP production in a wide variety of normal cell types [12-141, the expression of PTHrP in diverse tissues during development [15-17], and the presence of receptors recognizing PTHrP in many tissues [18,19] suggest that PTHrP has a role in normal growth, development, and differentiation in these cells and tissues. Recently, Iwamura et al. showed that 100%(33/33) of human prostatic adenocarcinomas contained cells that stained positive with an anti-PTHrP monoclonal antibody . This group later examined normal prostate and BPH . They demonstrated PTHrP expression only in neuroendocrine-like cells of these tissues , with no staining of glandular epithelium. In contrast, Kramer et al.  showed PTHrP immunoreactivity in the glandular epithelium of normal prostate. It has been known for some time that the human prostatic cancer cell line PC-3 secretes PTHrP into culture medium [23,24]. The human prostatic cancer cell lines DU-145 and LNCaP also show weak expression .However, no studies have characterized the expression of PTHrP in primary cultures of human prostatic cells or the regulation of PTHrP expression in prostatic cells. In this study we used immunohistochemistry to demonstrate PTHrP in the glandular epithelium of normal and BPH tissues, and in adenocarcinoma. 21 PTHrP expression in epithelial cell cultures derived from normal, BPH, and malignant tissues was analyzed by immunohistochemistry, northern analysis, and radioimmunoassay (RIA). Our results indicate that prostatic epithelial cells secrete abundant amounts of PTHrP and that epidermal growth factor (EGF) is a major inducer of PTHrP expression in these cells. 1,25(OH)2D,, implicated as an inhibitor of PTHrP expression in some other cell types, does not regulate PTHrP expression in cultured prostatic epithelial cells. MATERIALS AND METHODS Imrnunochemistry Tissue samples of -1 cc in size were dissected from radical prostatectomy specimens within 1 hr after surgery. Each sample was fixed in 10 volumes of Histochoice MB (Amresco, Solon, Ohio) for -24 hr at room temperature. The samples were then stored in 70% ethanol until they were processed and embedded in paraffin. Serial sections, 5 p m thick, were cut from each block and mounted on glass slides. One section was stained with hematoxylin and eosin, and examined by light microscopy to confirm the histological identity of the tissue. For immunohistochemical analysis, sections were deparaffinized with Hemo-de (Fisher, Pittsburgh, PA) and hydrated through serial baths of ethanol into phosphate-buffered saline. Endogenous peroxidase was blocked by immersion for 30 min in a solution of methanol with 0.3%hydrogen peroxide. The sections were then incubated overnight with a mouse monoclonal antibody, 9H7, against ITHrP 109-141  at 20 pg/ml. Staining was performed with biotinylated anti-mouse IgG and the ABC reagent (Vector Laboratories, Burlingame, CA) and developed with 3,3' diaminobenzidine. Slides were lightly counterstained with hematoxylin. Immunocytochemistry on cultured cells was performed as described for tissue, except that the cells were fixed in 2% paraformaldehyde for 15 min at room temperature and permeabilized in ice-cold 95%ethanol for 10 min. Cell Culture Establishment and maintenance of epithelia1 cell cultures from histologically normal peripheral zone, normal central zone, benign hyperplastic nodules (BPH), and cancers (Cap) of specimens obtained by radical prostatectomy were as previously described [26,27]. A small wedge of tissue was dissected from each specimen. The tissue was minced and digested overnight with collagenase. After rinsing and centrif- 22 Cramer et al. ugation to remove collagenase and most stromal cells, the digested tissue was inoculated into a 60-mm tissue culture dish coated with collagen type I and containing medium PFMR-4A supplemented with growth factors and hormones . Cells that grew out in primary culture were aliquoted and stored frozen in liquid nitrogen. Secondary cultures derived from the frozen aliquots were grown in MCDB 105 (Sigma, St. Louis, MO) supplemented with growth factors and hormones  until switching to experimental media (see later). The epithelial nature of these cells was verified by immunocytochemical staining of keratin and prostate-specific antigen. The secondary cultures are identical to the primary cultures in the pattern of expression of these markers, suggesting that they represent a similar population of cells. To venfy the histology of origin, the prostate was inked after dissection, fixed, and serially sectioned. The histology of sections immediately adjacent to and surrounding the portion removed for culture was reviewed. Experimental Media Complete medium for epithelial cells is defined as PFMR-4A  supplemented with cholera toxin (CT; 10 ng/ml), EGF (10 ng/ml), bovine pituitary extract (BPE; 40 pg/ml), insulin (IN; 4 ~g/ml),hydrocortisone (HC; 1 pg/ml), phosphoethanolamine (0.1 mM), selenous acid (3 x lo-' M), alpha-tocopherol (2.3 x M), retinoic acid (3 x lo-" M), and gentamicin (100 pg/ml). Basal medium for epithelial cells is defined as PFMR-4A supplemented with phosphoethanolamine, selenous acid, alpha-tocopherol, retinoic acid, and gentamicin at the levels indicated earlier for complete medium. RNA Analysis Total RNA was isolated from cultured cells by the guanidinium-phenol-chloroform extraction method . Ten micrograms of total RNA was fractionated on 1%agarose/6% formaldehyde slab gels and transferred to Hybond nylon membranes (Amersham, Arlington Heights, IL) using standard protocols (301. An RNA ladder (Bethesda Research Laboratories, Gaithersburg, MD: 0.24-9.5 kb) was run in a parallel lane as a size marker. Hybridization was carried out in 5% sodium dodecyl sulfate (SDS), 50 mM piperazine-N'-N'-bis(2ethanesulfonic acid), 100 mM NaCl, 50 mM sodium phosphate, 1 mM ethylenediaminetetraacetic acid (EDTA), pH 6.5, containing 1-2 X lo6 cpdml of the 1.6 kb EcoRI PTHrP cDNA fragment derived from the human renal carcinoma cell line 786-0 , radio- labeled using the Random Priming System I from New England Biolabs (Beverley, MA). Hybridization was for 16-18 hr at 60°C. After hybridization, the membranes were washed twice with a solution of 5% SDS and 1X SSC (150 mM NaC1, 15 mM sodium citrate, pH 7.0) at room temperature for 15 min each, followed by two washes in the same solution at 60°C for 30 min each, and a final quick rinse in 2 x SSC at room temperature. The membranes were exposed on Kodak XAR film(Eastman Kodak, Rochester, NY) at -80°C for 18-40 hr. The blots were subsequently stripped of radiolabeled probe in 0.1% SDS, 2 mM EDTA, pH 8.0, at 80°C for 10 min. Stripped blots were probed for expression of the ribosomal protein gene L7 (0.9 kb human cDNA probe)  to control for RNA loading and transfer differences. In several cell systems L7 mRNA has been shown to remain constant during different hormonal treatments and growth conditions [33-351. Hybridization and washing conditions were identical to those used for PTHrP. Autoradiogram exposure times were 1-8 hr. Densitometric analysis of autoradiograms was performed on a Molecular Dynamics Imagequant model 300 A laser densitometer (Sunnyvale, CA) using the Imagequant software provided by the manufacturer. Radioimmunoassay of PTHrP in Culture Medium Conditioned media were collected at the indicated times after addition of experimental media. Cellular debris was removed by centrifugation at 2000 X g for 5 min. A protease inhibitor cocktail (final concentrations: chymostatin, 2 pg/ml; leupeptin, 0.5 pg/ml; soybean trypsin inhibitor I, 50 kg/ml; pepstatin, 1.4 pg/ml; benzamidine, 333 pg/ml) was added and the medium was snap-frozen on dry ice and stored at - 70°C until assay of PTHrP.Assay of PTHrP (1-34) in conditioned media was conducted as described previously . In brief, the immunoassay was conducted using a rabbit polyclonal antibody directed against human PTHrP 1-34 with the same peptide used as the tracer and standard. All samples were run in at least triplicate. The detection limit of the assay was -20 pg/ml. The inter- and intra-assay coefficients of variation were 7% and 12%, respectively. Statistics Statistical analysis of EGF-stimulated PTHrP secretion was performed using Statview version 3.0, Abacus Concepts (Berkeley, CA). Data were analyzed by ANOVA followed by Scheffe's F-test. A P value I 0.05 was considered sigruficant. PTHrP in Human Prostatic Cells A BPH B 23 Cancer D C Fig. 1. Immunohistochemistry. Tissue sections and immunohistochemistry were as described in Materials and Methods. A BPH, 200X; B Cap, 200X; C: BPH, 8 0 0 ~ and ; D Cap, 8 0 0 ~ . RESULTS lmmunohistochemical Detection of PTHrP Using a monoclonal antibody against human PTHrP amino acids 109-141 (9H7), we labeled prostatic tissues and cell cultures with an immunoperoxidase technique. Normal, BPH, and malignant specimens derived from radical protatectomies were fixed with Histochoice MB, a fixative that does not contain formaldehyde or alcohol, and therefore leads to maximum preservation of antigenicity. In the normal and BPH specimens, we noted intense labeling of the glandular epithelium. This labeling was more or less uniform throughout the epithelium of the tissue sections. Malignant cells of adenocarcinomas were also labeled intensely by anti-PTHrP antibody. Figure 1 shows representative staining of BPH and adenocarcinoma specimens. Epithelial cell cultures derived from normal and malignant tissues were also labeled intensely and uniformly by antibody 9H7 (not shown). Although not apparent in the sections shown in Figure 1, diffuse staining by anitbody 9H7 of secreted material present in glandular lumens was observed (not shown). Expression of Multiple Transcripts by Cultured Epithelial Cells We next investigated PTHrP gene expression in human prostatic epithelial cell cultures. Total RNA for northern analysis was isolated from cultures of epithelial cells derived from normal, BPH, or malignant tissues. Figure 2 illustrates the expression of multiple PTHrP RNA species in all epithelial cell strains examined. Predominant bands of 1.5 kb, 2.1 kb, and 3.2 kb are indicated, although other minor bands were also present. To control for RNA loading, the same northern blot was stripped and reprobed for expression of the ribosomal protein gene L7 (Fig. 2, panel labeled L7). There was no apparent pattern in the level of PTI-IrP expression with regard to histology of the tissue of origin (e.g., normal, BPH, or CaP), although the level of expression was variable. We also examined PTHrP (1-34) secretion into culture media by RIA. PTHrP was secreted into culture media from all epithelial-derived cell strains tested (n > 10 for strains derived from normal, BPH, and malignant tissues). Values ranged from 0.5 to 5 ng/106 cells172 hr. 24 9.5 7.5 4.4 Cramer et al. - - 1.4 - .24 - 2.4 PTHI-P A -- PTHW C- 3.2 kb + 3.2 Y C- 2.1 kb 2.1 kb C.1.5 kb 1.5 kb L7 C 1.0 kb Time (hr) I I I I I I I I I 0 0.5 1 2 4 8 24 48 72 Fig. 2. PTHrP transcripts in epithelial cells. RNA samples from cultured prostatic cells derived from tissues of the indicated histologies were used t o prepare northern blots, as described in Materials and Methods. Blots were hybridized to radiolabeled PTHrP cDNA. The same blots were later stripped and reprobed with radiolabeled L7 cDNA t o control for loading. Lanes labeled C contain RNA from an epithelial cell strain (E-BPH-I) used as an internal control. Each lane was loaded with RNA from a different strain of the indicated histology. Limes above STD (kb) represent RNA standard migration run in a separate lane on the same gel, transferred to nylon membrane, and stained with methylene blue. Numbers next to each line correspond t o the size in kilobase pairs of the RNA standard band. Time (hours) Upregulation of PTHrP mRNA and Protein Expression in Epithelial Cells by Fresh Complete Medium Epithelial cells derived from BPH (strain E-BPH-1) were grown to semiconfluency and allowed to remain for 96 hr without a medium change ("starved"). "Starved" cells were then given fresh complete medium at TO, and RNA was isolated from parallel cultures at 0.5, l, 2, 4, 8, 24, 48, and 72 hr. A northern followed blot was prepared and probed for I", by stripping and reprobing for the ribosomal protein gene L7 to control for RNA loading and transfer differences (Fig. 3A). Figure 38 shows the densitometry data with PTHrP normalized to L7. PTHrP transcripts were detected by 1hr and continued to rise, reaching a peak by 24 hr, after which levels declined toward baseline values. The 2.1 kb transcript was the most abundant message, followed by the 1.5 kb transcript, and then the 3.2 kb transcript. AU transcripts responded to fresh complete medium with a similar magnitude of induction over baseline values (=lofold increase at 24 hr). Secretion of PTHrP into culture media by the cell strain E-BPH-1 and another cell strain derived from CaP (E-CA-I) was measured by RIA (Fig. 3C). Cells were grown in 35-mm culture dishes to semiconfluency and starved for 96 hr. Fresh complete medium w = 30 0 0.5 1 2 4 8 24 48 72 Time (Hours) Fig. 3. Time course of PTHrP expression. A Northern analysis. Epithelial cells (strain E-BPH- I) were grown t o semiconfluency in complete medium. The cultures were allowed t o go 96 hr without a medium change before feeding fresh medium (TO). RNA was isolated at the indicated time points, and 10 p.g was subjected to northern analysis as described in Materials and Methods. 8: Densitometry of autoradiograms from A. C Radioimmunoassay of PTHrP in conditioned media from Cap- (strain E-CA- I) and BPHderived (strain E-BPH-I) cells grown under similar conditions as those in A. After harvest of media, as described in Materials and Methods, cells were harvested by trypsinization and counted in a hemytometer. Results for E-CA-I and E-BPH-I are from two separate experiments. Each point is the mean of two separate plates. TO represents unconditioned medium that is below the detectable limit of the assay. was added at TO and followed by medium collection at 0.5, 1, 2, 4, 8, 24, 48, and 72 hr. Both cell strains showed increasing accumulation of PTHrP in the medium beginning at 4 hr and continuing up to 72 hr. PTHrP in Human Prostatic Cells - Regulation of PTHrP Transcript Levels by EGF We next investigated several components in the culture medium that might be responsible for the PTHrP induction noted after addition of fresh complete medium. Epithelial cells from normal prostate (E-PZ-l), BPH (E-BPH-I), and CaP (E-CA-1) were grown to semiconfluency and then starved for 96 hr. Fresh medium was added in different formulations, followed by total RNA isolation 24 hr later. The medium formulations were as follows: complete medium (see Materials and Methods); complete medium with CT, EGF, BPE, HC, or IN individually removed; basal medium (see Materials and Methods); basal medium with CT, EGF, BPE, HC, or IN individually added. All factors were present in the concentrations described in Materials and Methods. Figure 4A shows autoradiograms from a northern blot prepared from RNA isolated from strain E-BPH-1. Figure 4B shows the densitometric data with PTHrP normalized to the ribosomal protein gene L7. PTHrP expression in starved cells (TO) was barely detectable at the exposure shown. As demonstrated in the time course experiment described in the previous section, addition of complete medium resulted in increased PTHrP transcript levels at 24 hr. Individual removal of CT,BPE, HC, or IN from complete medium had no consistent effects on PIXrP transcripts. However, removal of EGF from complete medium resulted in =5- to 6-fold reduction in PTHrP transcript levels compared with complete medium. This effect was consistent for cells derived from normal, BPH, and CaP tissues (normal and CaP data not shown). Conversely, addition of EGF to basal medium resulted in an =4fold increase in PTHrP transcript levels over basal conditions. Regulation of PTHrP Secretion by EGF We next tested the effects of EGF on PTHrP secretion into culture medium. Four normal, one BPH, and one Cap-derived epithelial cell strains were tested for PTHrP (1-34) secretion by RIA.Cells were grown in complete medium to semiconfluency and starved for 72-96 hr, then complete medium +-EGF was added. After 24 hr, media were harvested and assayed for PTHrP by RIA, as described in Materials and Methods. Data are plotted in Figure 5, with cell strains grouped according to histology of origin (results are similar if all the data are pooled regardless of histology). Removal of EGF from complete medium resulted in a s i e c a n t reduction of PTHrP secretion into culture media. The decrease in PTHrP secretion ranged from 35% to 44% and was significant to P 5 0.001 for normal (n = 10) and BPH (n = 7). Statistics were not conducted on Cap-derived cells since we 25 PTHrP 3.2 kb 2.1 kb 1.5 kb L7 f- 1.0 kb B3 2.lkb 3.2kb 2 0 U Complete Basal Fig. 4. EGF regulation of PTHrP transcript levels. A: Northern analysis. Epithelial cells (strain E-BPH- I) were grown to semiconfluency in complete medium. The cultures were allowed to go 72 hr without a medium change before feeding fresh medium (TO). Experimental medium was either complete medium minus the indicated factor (lanes over solid line labeled “Complete”), or basal medium plus the indicated factor (lanes over solid line labeled “Basal”). Basal and complete media are defined in Materials and Methods. One plate remained in old medium until RNA isolation (Starved). After 24 hr RNA was isolated and 10 pg was analyzed, as described in Materials and Methods. B Densitometry of autoradiograms from A. tested one strain only once (i.e., n = l), but the results followed the same pattern as those for normal and BPH-derived cells. In similar experiments we have demonstrated that EGF stimulates PTHrP secretion into basal medium (data not shown). I,25(OHhD, Did Not Regulate PTHrP Transcript Levels Having shown that EGF could positively regulate PTHrP transcript levels, we chose to investigate the effects of 1,25(OH),D3 on PTHrP transcript levels. Inhibitory effects of 1,25(OH),D3 on PTHrP transcript levels in many cell types have been reported [37-401. Strain E-BPH-1 was treated with complete medium 26 Crarner et al. A -- PTHrP 3.2 kb 2.1 kb 1.5 kb L7 1.0 kb 1 Noml BPH Cancer Fig. 5. EGF increased production of PTHrP. Secretion of PTHrP into conditioned culture medium was measured by radioimmunoassay as described in Materials and Methods. Epithelial cell strains from tissues of the indicated histologies were grown to semiconfluency and starved for 72-96 hr before being switched to complete medium or complete medium minus EGF. Twenty-four hours later media were harvested and assayed for PTHrP as described in Materials and Methods. Data were normalized for each experimental pair using cells grown in complete medium as 100%. Each bar represents the mean *SE. *P 5 0.00 I. Time (hr) 1,25(OH)ZD3 ( 1 0 nM) 0 - + -1 1-1 1-1 1-1 1-1 1 - 2 4 + - - + 8 24 + - + 48 - + 1.5 kb+D t 2.1 kb+D 2 4 8 24 48 Time (hours) containing 10 nM of 1,25(OH),D, or vehicle (ethanol, 0.01%)and RNA was isolated after 2,4, 8, 24, and 48 hr. Figure 6A shows the autoradiographic results from a northern blot probed for PT'HrP expression that was stripped and reprobed for ribosomal protein gene L7 expression. Figure 6B shows the densitometry of the autoradiograms shown in Figure 6A. No major effects of 1,25(OH),D, could be seen at any time point for all three major PTHrP transcripts. A similar lack of effect of 1,25(OH),D3 on PTHrP transcripts was obtained when a Cap-derived cell strain (E-CA-1) was tested (data not shown). We have also tested the effects of 1,25(OH),D, on PTHrP expression in basal medium to remove any potential masking effects of EGF. No effects of 1,25(OH),D, could be seen (data not shown). A positive marker of 1,25(OH),D, action in prostatic cells is the induction of 1,25-dihydroxyvitamin D3-24hydroxylase transcripts (24hydroxylase) . We found that 24-hydroxylase transcripts were induced in the samples treated with 1,25(OH),D3 but not in vehicle controls (data not shown). DISCUSSION Previous reports have shown PTHrP expression in human prostatic tissues [20-221. In studies by Iwamura et al., PTHrP expression was abundant in prostatic adenocarcinoma , whereas PTHrP expression in benign prostatic tissue (normal prostate and BPH) was localized to the neuroendocrine cells . These studies used two different monoclonal anti- Fig. 6. 1,25-Dihydroxyvitamin D3 did not regulate PTHrP. A Northern analysis. Epithelial cells (stain E-BPH-I) were grown t o semiconfluency in complete medium. The cultures were allowed t o go 72 hr without a medium change before feeding fresh medium (TO). Experimental medium was complete medium I 0 nM I,25dihydroxyvitamin D3 [ 1,25(OH)ZD3]. RNA was isolated at the indicated times, and 10 fig was analyzed as described in Materials and Methods. 6: Densitometry of autoradiograms from A. * bodies raised against different epitopes of human PTHrP. In studies with benign tissues, monoclonal antibody 8B12 raised against PTHrP 1-34 was used , while in studies with cancer tissue, monoclonal antibody 9H7 raised against PTHrP 109-141 was used . Our immunohistochemical studies demonstrating FTHrP expression in glandular epithelium of normal prostate and BPH, and in prostatic cancer, utilized monoclonal antibody 9H7. For benign tissues, the differences in our results from Iwamura et al. may be due to the differences in the epitopes recognized by the two antibodies used, or in tissue fixatives and processing. It is also possible that proteolytic processing of PTHrP results in selective expression of epitopes of PTHrP in different cell types within the prostate. Kramer et al., using a panel of monoclonal and polyclonal antibodies against various regions of PTHrP (inclusive of the epitopes recognized by 8B12 and 9H7), have reported expression of PTHrP in normal prostatic glandular epithelia . Our results are PTHrP in Human Prostatic Cells consistent with those of Kramer et al.  demonstrating that glandular epithelial cells express PTHrP. We also demonstrated the expression of PTHrP transcripts in cultured epithelial cells from normal prostate, BPH, and Cap. Every strain tested expressed multiple transcripts for PTHrP. To our knowledge, no previous study has investigated PTHrP expression in primary cultures of prostatic cells. Previous immunohistochemical analysis of the pattern of expression of keratins, prostate-specific antigen, and prostatic acid phosphatase indicate that our cultures are more like glandular epithelial cells than neuroendocrine cells . The results reported here using cultured cells are consistent with our findings in prostatic tissue sections. We investigated the factors in serum-free culture medium affecting PTHrP expression in prostatic epithelial cells. EGF was the major regulator of PTHrP transcripts. Studies using other cell lines have shown the effects of EGF on PTHrP expression [38,40,43]. Interestingly, with human keratinocytes, Kremer et al.  found EGF effective only in combination with BPE. However, in prostatic cells we find that EGF regulates PTHrP expression in the absence of BPE. We do not believe that the effects of EGF on PTHrP expression are related to the growth-stimulatory effects of EGF on prostatic cells. In support of this, all of the other factors in serum-free medium we tested have growth-stimulatory properties in our culture system but did not regulate PTHrP expression. Previous studies have shown mixed effects of 1,25(OH),D3on PTHrP expression. Positive [MI, negative [37-401, and no effects  of 1,25(OH),D3 on PTHrP production have been reported. We found no effect of 1,25(OH),D3 on PTHrP expression in primary human prostatic epithelial cells. Our laboratory has recently demonstrated the presence of vitamin D receptors in cultures of primary human prostatic cells , and we  and others [47l have characterized vitamin D receptors in several human prostatic cancer cell lines. We have also demonstrated antiproliferative effects of 1,25(OH),D3 on primary human prostatic cells  and human prostatic cancer cell lines . In addition, the cells used in this study exhibited a normal response of 1,25(OH),D3 induction of 24hydroxylase. The PTHrP gene has three promoters and multiple exons in the 5' untranslated region of the gene [48,49]. Cell typespecific differences in 1,25(OH),D3 regulation of PTHrP gene expression could be due to tissue-specific transcription factors and/or tissue-specific promoter utilization [48,50]. It is possible that normal stromaYepithelia1interactions are required for 1,25(OH),D3 regulation of PTHrP expression in the prostate; however, this requirement has not been shown for other cell types [37-40,441. 27 CONCLUSIONS The significance of PTHrP production by the prostate is not known. Recently, M'HrP has been shown to increase 3H-thymidine incorporation in several prostatic epithelial cell lines, suggesting an autocrine growth-promoting role for PTHrP in the prostate . PTHrP has also been shown to relax smooth muscle cells [51,52], and smooth muscle cells can be found throughout the stromal component of the prostate. The data presented in the present study, demonstrating epithelial production of PTHrP, suggests that autocrine and/or paracrine effects of PTHrP may be important. The identification of factors produced by prostatic cells that have known biological effects on bone could be important for understanding the mechanism of prostatic metastasis to bone. The hypercalcemic effects of PTHrP are thought to be mediated by the N-terminal region of the PTHrP molecule acting on the PTWPTHrP receptors on bone and kidney cells . However, C-terminal regions of the PTHrP protein have been implicated in an inhibition of osteoclastic activity . The net effect of PTHrP on bone remodeling could be determined by the nature of the PTHrP peptides secreted. Elucidation of the PTHrP peptides secreted by the prostate, and characterization of the bioactivity of these fragments on bone, would further our understanding of the role of PTHrP in prostate-mediated bone remodeling. It is conceivable that PTHrP may be involved in creating a suitable microenvironment for prostatic cell growth within the bone matrix. In the future, studies directed at understanding autocrine/paracrine effects of PTHrP on prostatic growth and differentiation, and the role of PTHrP in prostatic bone metastasis and prostate-mediated bone remodeling, are merited. ACKNOWLEDGMENTS We thank Gordon Strewler for generously providing the PTHrP cDNA and Judith Campisi for providing the L7 cDNA. This work was supported by grants from the Lucas Foundation, Menlo Park, CA, the CaP CURE Foundation, Santa Monica, CA, and NIH grant DK47551-02 to D.M.P.; the American Institute for Cancer Research grant 92b29 and NIH grant DK42482 to D.F.; National Research Service Award CA59086 from the National Cancer Institute of the NIH to S.D.C.; and National Institutes of Health (CA71347) and Department of Veterans Affairs grants to L.J.D. REFERENCES 1. Boring CC, Squires TS, Tong T, Montgomery S: Cancer statistics. CA Cancer J Clin 44:7-26, 1994. 28 Crameret al. 2. Galasko CS: The anatomy and pathways of skeletal metastasis. In Weiss L, Gilbert HA (eds): “Bone Metastasis.” Boston: G.K. Hall Medical, 1981, pp 49-63. 3. Franks LM:Etiology, epidemiology and pathology of prostatic cancer. Cancer 32:1092-1095, 1973. 4. Yoshida K, Akimoto M: Computed tomographic evaluation of bone metastases in prostatic cancer patients. In Karr J, Yamanaka H (eds): “Prostate Cancer and Bone Metastasis.” New York: Plenum Press, 1992, pp 197-204. 5. Suzuki T, Shimizu T, Kurokawa K, Jimbo H, Sat0 J, Yamanaka H: Pattern of prostate metastasis to the vertebral column. Prostate 25:141-146, 1994. 6. Stewart AF: Humoral hypercalcemia of malignancy. In Favus MJ (ed): “Primer on the Metabolic Diseases and Disorders of Mineral Metabolism.” 2nd ed. New York: Raven Press, 1993, pp 169-173. 7. Abou-Samra A, Juppner H, Force T, Freeman M W , Kong X, Schipani E, Urena P, Richards J, Bonventre JV, Potts JTJ, Kronenberg HM, Segre GV: Expression cloning of a common receptor for parathyroid hormone and parathyroid hormone-related peptide from rat osteoblast-like cells: A single receptor stimulates intracellular accumulation of both CAMPand inositol trisphosphates and increases intracellular free calcium. Proc Natl Acad Sci USA 89~2732-2736,1992. 8. Lee K, Deeds JD, Chiba S, Un-No M, Bond AT, Segre G: Parathyroid hormone induces sequential c-fos expression in bone cells in vivo: In situ localization of its receptor and c-fos messenger ribonucleic acids. Endocrinology 134:441-450,1994. 9. Karaplis AC, Luz A, Glowacki J, Bronson RT, Tybulewicz VL, Kronenberg HM, Mulligan RC: Lethal skeletal dysplasia from targeted disruption of the parathyroid hormone-related peptide gene. Genes Dev 8~277-289,1994. 10. Bouizar Z, Spyratos F, Deytieux S, de Vernejoul MC, Jullienne A Polymerase chain reaction analysis of parathyroid hormone-related protein gene expression in breast cancer patients and occurrence of bone metastases. Cancer Res 53976-5078, 1993. 11. Guise TA, Taylor SD, Yoneda T, Sasaki A, Wright K, Boyce BF, Chirgwin JM,Mundy GR: Parathyroid hormone-related protein (PTHrP) expression by breast cancer cells enhance osteolytic bone metastasis in vivo. Sixteenth Annual Meeting of the American Society of Bone an Mineral Research. 9(Suppl 1):S128 #30, 1994. 12. Hayman JA, Danks JA, Ebeling PR, Moseley JM, Kemp BE, Martin TJ: Expression of parathyroid hormone related protein in normal skin and in tumors of skin and appendages. J Pathol 158:293-296, 1989. 13. Asa S1, Henderson J, Goltzman D, Drucker DJ: Parathyroid hormone-like peptide in normal and neoplastic human endocrine tissues. J Clin Endocrinol Metab 71: 1112-1118, 1990. 14. Selvanayagam P, Graves K, Cooper C, Rajaraman S: Expression of parathyroid hormone-related peptide gene in rat tissues. Lab Invest 64:713-717, 1991. 15. Burton PBJ, Moniz C, Quirke P, Tzannatos C, Pickles A, Dixit M, Triffit JT, Jeuppner H, Segre GV, Knight DE: Parathyroid hormone-related peptide in the human fetal uro-genital tract. Mol Cell Endocrinol69:R13R17, 1990. 16. Moseley JM, Hayman JA, Danks JA, Alcorn D, Grill V, Southby J, Horton MA: Immunohistochemical detec- tion of parathyroid hormone related protein in human fetal epithelia. J Clin Endocrinol Metab 73:478-484, 1991. 17. Schermer DT, Chan SDH, Bruce R, Nissenson RA, Wood WI, Strewler GJ: Chicken parathyroid hormonerelated protein and its expression during embryologic development. J Bone Miner Res 6:149-155, 1991. 18. Tian J, Smorgorzewski M, Kedes L, Massry SG: Parathyroid hormone-parathyroid hormone related protein receptor messenger RNA is present in many tissues beside the kidney. Am J Nephrol 13:210-213, 1993. 19. Urena P, Kong XF, Abou-Samra AB, Jeuppner H, Kronenberg HM, Potts JT, Segre GV PTH/PTHrP receptor mRNA’s are widely distributed in rat tissues. Endocrinology 133:617-623, 1993. 20. Iwamura M, di Sant’Agnese PA, W u G, Benning CM, Cockett ABT, Deftos LJ, Abrahamsson P: Immunohistochemical localization of parathyroid hormone-related protein in human prostate. Cancer Res 53:1724-1726, 1993. 21. Iwamura M, Wu G, Abrahamsson P, Di SanYagnese PA, Cockett ATK, Deftos LJ: Parathyroid hormone-related protein is expressed by prostatic neuroendocrine cells. Urology 43567-674, 1994. 22. Kramer S, Reynolds FH, Castillo M, Valenzuela DM, Thorikay M, Sorvillo JM: Immunological identification and distribution of parathyroid hormone-related protein polypeptides in normal and malignant tissues. Endocrinology 128~1927-1937,1991. 23. Rodan SB, Insogna KL, Vignery AMC, Stewart AF, Broadus AE, D’Souza SM, Bertolini DR, Mundy GR, Rodan GA: Factors associated with humoral hypercalcemia of malignancy stimulate adenylate cyclase in osteoblastic cells. J Clin Invest 721511-1515, 1983. 24. Kao PC, Klee GG, Taylor RL, Heath H: Parathyroid hormone-related peptide in plasma of patients with hypercalcemia and malignant lesions. Mayo Clin Proc 65: 1399-1406, 1990. 25. Iwamura M, Abrahamsson P, Foss KA, Wu G, Cockett AT, Deftos LJ: Parathyroid hormone-related protein: A potential autocrine growth regulator in human prostate cancer cell lines. Urology 43:675-679, 1994. 26. Peehl DM: Culture of human prostatic epithelial cells. In Freshney IA (ed): “Culture of Epithelial Cells.” New York: Wiley Liss, 1992, pp 159-180. 27. Peehl DM, Skowronski RJ, Leung GK, Wong ST, Stamey TA, Feldman D: Antiproliferative effects of 1,25dihydroxyvitamin D3 on primary cultures of human prostatic cells. Cancer Res 54:805-810, 1994. 28. Peehl DM: Serial culture of adult human prostatic epithelial cells. J Tissue Cult Techn 9:53-60, 1985. 29. Chomuynski P, Sacchi N: Single-step method of RNA isolation by acid guanidium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156-159, 1987. 30. Ausubel F, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K “Current Protocols in Molecular Biology.” New York: Green and Wiley Interscience, 1989. 31. Virca GD, Northemann W, Shiels BR, Widera G, Bromme S: Simplified northern blot hybridization using 5% sodium dodecyl sulfate. Biotechniques 8:370371, 1990. 32. Thiede MA, Strewler GJ, Nissenson RA, Rosenblatt M, Rodan GA: Human renal carcinoma expresses two messages encoding a parathyroid hormone-like pep- PTHrP in Human Prostatic Cells tide: Evidence for the alternative splicing of a singlecopy gene. Proc Natl Acad Sci USA 85:4605-4609,1988. 33. Seshadra T, Campisi J: Regression of c-fos transcription and an altered genentic program in senescent human fibroblasts. Science 247205-209, 1990. 34. Krishnan A, Feldman D Activation of protein kinase-C inhibits vitamin D receptor gene expression. Mol Endocrinol 5~605-612, 1991. 35. Cramer SD, Barnard R, Engbers C, Ogren L, Talamantes F: Expression of the growth hormone receptor and growth hormone-binding protein during pregnancy in the mouse. Endocrinology 1312376-882, 1992. 36. Burton DW, Brandt DW, Deftos LJ: Parathyroid hormone related protein in the cardiovascular system. Endocrinology 134:253-261, 1994. 37. Ikeda K, Lu C, Weir EC, Mangin M, Broadus AE: Transcriptional regulation of parathyroid hormone-related peptide gene by glucocorticoids and vitamin D in a human C-cell line. J Biol Chem 264:15743-15746, 1989. 38. Kremer R, Karapalis AC, Henderson J, Gulliver W, Banville D, Hendy GN, Goltzman D: Regulation of parathyroid hormone-like peptide in cultured normal human keratinocytes. J Clin Invest 87884-893, 1991. 39. Haq M, Kremer R, Goltzman D, Rabbani SA: A vitamin D analog (EB1089) inhibits parathyroid hormone-related peptide production and prevents the development of malignancy-associated hypercalcemia in vivo. J Clin Invest 91:2416-2422, 1993. 40. Liu 8, Goltzman D, Rabbani S A Regulation of parathyroid hormone-related peptide production in vitro by the rat hypercalcemic Leydig cell tumor H-500. Endocrinology 132:1658-1664, 1993. 41. Skowronski RJ, Peehl DM, Feldman D: Vitamin D and prostate cancer: 1,25 dihydroxyvitamin D3 receptors and actions in human prostate cancer cell lines. Endocrinology 132:1952-1960, 1993. 42. Peehl DM, Leung G, Wong S: Keratin expression: A measurement of phenotypic modulation of human prostatic epithelial cells by growth inhibitory factors. Cell Tissue Res 27731-18, 1994. 43. Rodan SB, Wesoloski G, Inacone J, Theide MA, Rodan GA: Production of parathyroid hormone-like peptide in a human osteosarcoma cell line: Stimulationby phorbol esters and epidermal growth factor. J Endocrinol 122: 219-227, 1989. 44. Merryman JI, Capen CC, McCauley LK, Werkmeister 45. 46. 47. 48. 49. 50. 51. 52. 53. 29 JR, Suter MM, Rosol TJ: Regulation of parathyroid hormone-related protein production by a squamous carcinoma cell line In vitro. Lab Invest 6937-354, 1993. Werkmeister JR, Merryman JI, McCauley LK, Horton JE, Capen CC, Rosol TJ: Parathyroid hormone-related protein production by normal human keratinocytes in vitro. Exp Cell Res 208:68-74, 1993. Peehl DM, Skowronski RJ, Leung GK, Wong ST, Stamey TA, Feldman D: Antiproliferative effects of 1,25-dihydroxyvitamin D3 on primary cultures of human prostatic cells. Cancer Res 545305-810, 1994. Miller GJ, Stapleton GE, Ferrara JA, Lucia MS, Pfister 5, Hedlund TE, Upandhya P: The human prostate carcinoma line LNCaP expresses biologically active, specific receptors for 1&5-dihydroxyvitamin D3. Cancer Res 52~515-520,1992. Brandt DW, Bruns ME, Bruns DE, Freguson JEI, Burton DW, Deftos LJ: The parathyroid hormone-related protein (PTHrP) gene preferentially utilizes a GC-rich promotor and the PTHrP 1-139 coding pathway in normal human amnion. Biochem Biophys Res Commun 189: 938-943, 1992. Vasavada RC, Wysolmerski JJ, Broadus AE, Philbrick WM: Identification and characterization of a GC-rich promoter of the human parathyroid hormone-related peptide gene. Mol Endocrinol7273-282, 1993. Campos RV, Wang C, Drucker DJ: Regulation of parathyroid hormone-related peptide (ITHrP) gene transcription: Cell- and tissue-specific promotor utilization mediated by multiple positive and negative cis-acting DNA elements. Mol Endocrinol 6:1642-1652, 1992. Winquist R, Baskin E, Vlasuk G: Synthetic tumor-derived human hypercalcemic factor exhibits parathyroid hormone-like vasorelaxation in renal arteries. Biochem Biophys Res Commun 149527-232, 1987. Okano K, Wu S, Huang X, Pirola CJ, Juppner H, AbouSamra AB, Segre GV, Iwasaki K, Fagin JA, Clemens TL: Parathyroid hormone (PTH)/FTH-related protein (ITHrP) receptor and its messenger ribonucleic acid in rat aortic vascular smooth muscle cells and UMR osteoblast-like cells: Cell-specific regulation by angiotensin-I1 and PTHrP. Endocrinology 135:1093-1099, 1994. Orloff JJ, Reddy D, De Papp AE, Yang KH, Soifer NE, Stewert AF: Parathyroid hormone-related protein as a prohormone: Posttranslational processing and receptor interactions. Endocrin Rev 15:40-60, 1994.