The Prostate 19:221-235 (1991)
Reconstituted Basement Membrane Promotes Morphological and Functional Differentiation of Primary Human Prostatic Epithelial Cells Chau-Jye Fong, Edward R. Sherwood, Debra M. Sutkowski, Graziella M. Abu-Jawdeh, Hidejiro Yokoo, Kenneth D. Bauer, James M. Kozlowski, and Chung Lee Departments of Urology (C.-J.F., E.R.S., D.M.S., J.M.K., C.L.), and Pathology (G.M.A.-J., H. Y., K.D.B.), Northwestern University Medical School, Chicago, Illinois Prostatic epithelial cells undergo rapid proliferation and lose their ability to synthesize and secrete prostate-specific antigen (PSA) and prostatic acid phosphatase (PAP) under standard tissue culture conditions. Herein, we compared the morphology, growth, secretory activity, and intermediate filament expression of human prostatic epithelial cells cultured on either standard tissue culture plastic or reconstituted basement membrane. Epithelial cells grown on plastic exhibited a 10-fold increase in proliferation and a higher percentage of cells in the S-phase of the cell cycle compared to cells cultured on basement membrane. However, cells grown on basement membrane secreted markedly higher levels of PSA and PAP. The basement membrane-induced enhancement of secretory activity was potentiated by dihydrotestosterone (DHT) and prostate stromal cell conditioned medium. Morphological studies showed that cells plated on basement membrane formed organoid-like clusters and maintained several aspects of differentiated epithelium including abundant secretory vesicles, microvilli, and desmosomes with associated cytoskeletal elements. Cultivation of epithelial cells on basement membrane components also suppressed the expression of vimentin, a mesenchymal intermediate filament polypeptide. However, cytokeratin expression was abnormal in cells grown on either surface. These results indicate that the differentiated properties of prostatic epithelial cells are promoted by cultivation on reconstituted basement membrane in the presence of DHT and stromal cell conditioned medium.
Key words: PSA, PAP, intermediate filaments
INTRODUCTION The mature secretory epithelium of the human prostate consists of highly differentiated cells that secrete a complex prostatic fluid. In vivo, the prostatic epithelium resides in an environment in which factors such as androgens, extracellular matrix, and stroma influence cellular growth and differentiation [ 1-41. In conventional culture, prostatic epithelial cells undergo a few rounds of cell division and rapidly lose differentiated function. Thus, their characteristics in vitro may have little relevance to
Received for publication April 2, 1991; accepted June 10, 1991 Address reprint requests to Chau-lye Fong, Department of Urology, Northwestern University Medical School, 303 East Chicago Avenue, Chicago, IL 60611.
0 1991 Wiley-Liss, Inc.
222
Fong et al.
the in vivo condition. This observation is further reflected in the inability of prostatic epithelial cells to synthesize and secrete prostatic acid phosphatase (PAP) and prostate-specific antigen (PSA) after cultivation on tissue culture plastic [5-61. Therefore, in order to effectively study prostatic epithelial cell biology, there is a need to establish culture conditions which maintain the normal functional characteristics of the prostatic epithelium. In previous studies, we reported techniques for the isolation and selective cultivation of human prostatic epithelial cells [5-6). In the present study, we attempted to define the role of basement membrane in the regulation of prostatic epithelial cell function. Specifically, we studied the cellular proliferation, morphology, secretory activity, and intermediate filament expression of prostatic epithelial cells cultured on either plastic or reconstituted basement membrane. We also assessed the impact of androgen and stromal cell conditioned medium on the secretory activity of primary prostatic epithelial cell cultures. MATERIALS AND METHODS Isolation of Epithelial and Stromal Cells From the Human Prostate
Prostatic tissue was obtained from individuals undergoing suprapubic or retropubic prostatectomy for treatment of bladder neck obstruction secondary to benign prostatic hyperplasia (BPH). Representative portims of the specimens were submitted for histopathologic evaluation to assure the diagnosis of BPH without focal prostatic carcinoma. The remaining tissue was minced into 1 mm3 fragments and transferred to sterile dissociation flasks containing RPMI- 1640 medium supplemented with 10%fetal bovine serum, 200 U/ml of collagenase type I (Sigma Chemical, St. Louis), and 100 pg/ml of DNAse type I (Sigma). The tissue was dissociated for 16-18 hours at 37°C using a magnetic stimng bar to provide gentle agitation. Following the dissociation period, the resulting cell suspension was washed, resuspended in Hanks’ balanced salt solution (HBSS), and layered over discontinuous percoll (Pharmacia, Uppsala, Sweden) gradients for separation as previously described [5-61. Epithelial cell aggregates were aspirated from gradients, washed (2 x ) with HBSS, and utilized in our studies. In all studies, cells were plated using WAJC 404 medium supplemented with 10% fetal bovine serum (plating medium). Twentyfour hours after plating, the media were changed into serum-free WAJC 404 supplemented with ITS (insulin at 5 pg/ml, transfemn at 5 pg/ml, selenous acid at 5 ng/ml, Collaborative Research, Bedford, MA) epidermal growth factor (3 ng/ml, Sigma), bovine pituitary extract (30 pg/ml, Collaborative Research), prolactin (3 ng/ml, Sigma), cholera toxin (10 ng/ml, Sigma), polyvinyl pyrrolidone (2 mg/ml, Berhring Diagnostics, La Jolla, CA), and penicillin (100 U/ml)/streptomycin (100 pg/ml). Stromal cells were harvested from gradients, washed, and cultured in RPMI- 1640 medium containing 10%fetal bovine serum. Stromal cell conditioned medium was prepared by culturing confluent monolayers in RPMI-I640 containing ITS for 48 hours. Assessment of Prostatic Epithelial Cell Morphology
Isolated epithelial cells were harvested directly from percoll gradients and plated (2 X 105/well) onto untreated 24 well plates (Falcon-Becton Dickinson, Lincoln Park, NJ) or 24 well plates coated with reconstituted basement membrane (Ma-
Human Prostate Epithelial Cell Biology
223
trigel, Collaborative Research, Bedford, MA). Matrigel consists of laminin, type IV collagen, heparan sulfate proteoglycans, and small amounts of entactin and nidogen [7]. The following day (day 0), the media were changed into serum-free WAJC 404 medium. Cells were cultured for 6 days and light photomicroscopy was performed periodically using a phase contrast microscope to document morphology. In electron microscopy studies, cells were cultured on plastic or matrigel as outlined above. After 5-12 days of cultivation, cells were washed (3 X ) with HBSS and fixed with Millonig’s fixative for 30 minutes at 4°C. Cells were washed (3 X ) with HBSS, dehydrated by successive incubations with 25-100% ethanol, and embedded in Epon. Thin sections were cut perpendicular to the dish surface and doubly stained with lead citrate and uranyl acetate for analysis by transmission electron microscopy. Mitogenesis Assay
Prostatic epithelial cells (5 X lo4 cells/well) were plated onto plastic or matrigel-coated surfaces in 24 well plates as outlined above and allowed to adhere overnight (16-18 hours). The following day (day 0), the cultures were washed (3 X ) with HBSS and day 0 cell counts were performed to determine plating efficiency in each group. Cells were cultured for 6 days in serum-free WAJC 404 and counted on days 2, 4, and 6 to assess growth. Existing media were replaced with fresh media on days 2 and 4. Cellular viability was determined using 0.25% trypan blue after detaching the cells from the culture surfaces. In all experiments, cells were detached from matrigel-coated surfaces using dispase (Collaborative Research) and from plastic using 0.1% trypsin (Hazleton, Lenexa, KS). Viable cells were counted using a hemacytometer. In all studies, cellular viability was greater than 90%.
Flow Cytometry DNA content and the cell cycle distribution of prostatic epithelial cells grown on plastic or matrigel for 6 days were determined using flow cytometry. The cells were plated in low density (2 X 104/well),so that after 6 days of culture the cells were less than 80% confluent. The cells were harvested, washed (3 X ) with HBSS, and resuspended (2 x 106/ml)in RPMI 1640 medium supplemented with 10% FBS and 10% DMSO. Cellular DNA was stained with propidium iodide using the Triton-X method [8]. DNA content was analyzed by measuring red fluorescence on the an Epics 752 flow cytometer (Coulter Corp, Hialeah, FL). Cell cycle analysis was performed using Multicycle software (Phoenix Flow Systems, San Diego, CA) which uses the algorithm of Dean and Jett [9]. Diploid chronic lymphocytic leukemia (CLL) cells were utilized as controls for DNA content determinations. Determination of Prostate Specific Antigen (PSA) and Prostatic Acid Phosphatase (PAP) Secretion by Prostatic Epithelial Cells Prostatic epithelial cells (8 X lo5 cells/well) were plated in 24 well plates and allowed to adhere overnight. The following day, media were changed to serum-free WAJC 404 containing specific additives and cells were cultured for 10 days. Media were replaced every other day with fresh media containing appropriate additives. Conditioned media were harvested from each well on day 10, centrifuged (400g for 10 minutes) to remove residual cells, and stored at -80°C. The cells in each well
224
Fong et al.
were harvested and counted on a hemacytometer. PSA and PAP levels in conditioned media were determined using the Tandem-R PAP and PSA immunoradiometric assay (Hybritech, San Diego, CA). For PAP, 100 pl of conditioned medium was added to 12 x 75 mm plastic tubes followed by monoclonal anti-PAP IgG-coated beads and 100 p1of '251-labeled anti-PAP IgG tracer antibody. The samples were incubated for 4 hours, washed (2 X ) with phosphate-buffered saline (PBS) and counted in an LKB minigamma system. PSA was analyzed in an identical fashion except that 50 p1 of conditioned medium was used and monoclonal antibodies were directed against specific epitopes on the PSA molecule. The radioactivity was directly proportional to the concentration of PAP or PSA in the test sample, which was determined from a standard curve.
lmmunoprecipitation Prostatic epithelial cells (5 X 10' cells/well) were plated in 24 well plates using WAJC 404 with 10% FBS and allowed to adhere overnight. The following day, the media were changed into serum-free WAJC 404 and cells were cultured for 8 days with media being changed every other day. The conditioned media were harvested and centrifuged (400g for 10 minutes) to remove residual cells. PAP and PSA were immunoprecipitated from epithelial cell conditioned media using Protein A sepharose CL-6B beads (Pharmacia) which were washed (2 X ) with phosphate-buffered saline (PBS) and resuspended at a concentration of 100 mg/ml in PBS. Twenty microliters of anti-PSA or anti-PAP (Dako, Santa Barbara, CA) was added to the slurry and incubated (4°C) for 6 hours with constant agitation. The beads were washed (3 X ) with PBS to remove excess antibody and resuspended in PBS as above. Five hundred microliters of protein A sepharose slurry was added to 20 ml of prostatic epithelial conditioned medium. Samples were incubated for 16 hours with constant agitation, washed (3 X ) with PBS, and eluted from the beads with urea mix (9 M urea, LKB ampholyte pH 9-1 1, 2% mercaptoethanol, 4% NP-40) for two-dimensional electrophoresis. For comparison, PSA and PAP were identified in expressed prostatic fluid by western immunoblotting as outlined below. Two-Dimensional Electrophoresis The ISO-DALT system for two-dimensional electrophoresis was performed according to the method of Anderson and Anderson [10,11]. Briefly, the first dimension (the IS0 system), which separates proteins according to their isoelectric points, was carried out for 14,000 volt hours in 5% acrylamide containing 8 M urea, 2% total ampholytes with a pH range of 3-10, and 1.7% nonidet detergent. At the end of the isoelectric focusing, each gel was extruded from the elongated glass tube (17.8 x 0.15 cm) and equilibrated in a solution containing 138 mM Tris base, 10% glycerol, 2% sodium dodecyl sulfate (SDS), and bromphenol blue (pH 6.8). The second dimension (the DALT system) separated proteins according to their molecular weights. The slab gels consisted of a linear polyacrylamide gradient [9-I 81 containing 1% SDS. The electrophoresis buffer contained 24 mM Tris, 0.2 M glycine, and 1% SDS. Electrophoresis was carried out at a constant voltage of 100-150 volts for 16 hours. Protein in gels was detected by the silver staining method of Guevara and colleagues [ 121. Western immunoblotting was performed by transferring proteins to nitrocellulose (Bio-Rad, 0.45 p.m) at 120 volts for 2 hours using 25 mM Tris, 192
Human Prostate Epithelial Cell Biology
225
mM glycine, and 20% methanol transfer buffer. Blots were washed (3 x ) with distilled water, incubated with blocking buffer (Carnation non-fat dry milk) for 2 hours and with primary antibody overnight. The blots were washed (3 X ) with blocking buffer and incubated with horseradish peroxidase-conjugated anti-mouse IgG for 2 hours. The color reaction was induced using 4-chloro- l-naphthol substrate. Analysis of Intermediate Filament Expression
For analysis of cytokeratin expression, epithelial cells were cultured on plastic or matrigel for 6 days. Cells were harvested and washed (3 X ) with PBS. Cells were dissolved in urea mix and proteins were separated using two-dimensional electrophoresis. Cytokeratins were analyzed by western immunoblotting using monoclonal anti-cytokeratins 8.12 and 8.13 (Sigma Chemical). For comparison, the cytokeratin profiles of fresh prostatic tissue and epithelial cells isolated directly from percoll gradients were determined as previously described [ 13,141. Vimentin gene expression was assessed by immunocytochemical analysis of purified epithelial cell cultures grown on plastic or surfaces coated with the basement membrane components type IV collagen or laminin. Type IV collagen and laminin (Collaborative Research) -coated surfaces were prepared by addition of 0.5 ml of collagen or laminin solution (10 kg/ml) to 24 well plates. The wells were allowed to air dry under aseptic conditions, and epithelial cells ( 5 X 104/well) were plated onto respective wells and cultured for 3 days. Cells were fixed with 100% methanol, washed (3 x ) with PBS, and incubated (4°C) overnight with mouse anti-human vimentin (Sigma Chemical Co., Clone VIM 13.2). Cells were washed (3 X ) with PBS, and immunoreactivity was determined using the Vectastain (Vector Laboratories, Burlingame, CA) ABC immunoperoxidase system and diaminobenzidine as substrate.
RESULTS Morphology of Prostatic Epithelial Cells Grown on Plastic or Matrigel Cultivation of human prostatic epithelial cells on matrigel was associated with the adoption of a clustered morphology that was not expressed when cells were maintained on plastic as shown in Figure 1. Specifically, cells cultured on plastic exhibited monolayer growth and fusiform morphology. Epithelial cells cultured on plastic formed subconfluent monolayers after 2 days of cultivation which grew to confluence by day 6. In contrast, cells cultured on matrigel for 2 days exhibited a rounded morphology. Between days 2 and 6, cells which were initially randomly dispersed over the growth surface migrated to form organoid-like clusters. By day 6, well-formed clusters were completed and could be maintained for more than 3 weeks. Electron microscopic analysis of epithelial cells cultured on plastic or matrigel is shown in Figure 2. Analysis of cell clusters obtained following cultivation of epithelial cells on matrigel for 5 days showed the presence of several layers of epithelial cells which possessed numerous microvilli (Fig. 2A). Cells grown on matrigel for 12 days also exhibited abundant microvilli as well as numerous secretory vesicles and luminal structures containing secretory material (Fig. 2B). Cells cultured on plastic grew in monolayer and possessed few microvilli or secretory vesicles (Fig. 2C). The presence of differentiation-associated structures in epithelial cells cultured
226
Fong et al. PLASTIC
MATRIGEL
Fig. I . Phase contrast micrographs of prostatic epithelial cells cultured on plastic or matrigel. Cells were plated onto plastic or matrigel-coated surfaces and cultured for 6 days. Cells were photographed on days 2 and 6 to demonstrate cellular morphology. Arrows show clusters of epithelial cells observed following cultivation of cells on matrigel for 6 days. ~ 4 0 0 .
on matrigel is further demonstrated in Figure 3. The presence of luminal structures with associated microvilli was observed following cultivation of epithelial cells on matrigel for 12 days (Fig. 3A). Epithelial cells cultured on matrigel for 5 days also possessed desmosomes with associated cytoskeletal elements (Fig. 3B). Desmosomes were not observed in epithelial cells cultured on plastic. Numerous secretory vesicles and pseudopodial extensions into the reconstituted basement membrane were also observed following 12 days of culture on matrigel (Fig. 3C). Growth of Prostatic Epithelial Cells on Plastic or Matrigel The proliferation of prostatic epithelial cells on plastic and matrigel was determined using cell counts as outlined in Figure 4. Cells cultured on plastic exhibited a doubling time of 42 hours during the log phase of growth and showed a greater than 5-fold increase in cell number over the 6 day cultivation period. Cells cultured on matrigel did not exhibit significant proliferation during the 6 days of cultivation but maintained a constant population of viable cells during the culture period. How cytometry studies confirmed the enhanced growth of cells cultured on plastic. Cells grown on plastic had a higher number of cells in the S phase of the cell cycle compared to cells grown on basement membrane matrigel (15.7% versus 9.3%). Both cell populations exhibited a diploid DNA content as determined by comparison to diploid chronic lymphocytic leukemia cells.
Human Prostate Epithelial Cell Biology
227
Fig. 2. Electron micrographs of epithelial cells cultured on plastic or matrigel. Cells were cultured on plastic or matrigel-coated surfaces and prepared for electron microscopy. A: Epithelial cells cultured on matrigel for 5 days, X 2,000. B: Epithelial cells cultured on matrigel for 12 days, X 4,000. C: Epithelial cells cultured on plastic for 5 days, X 2,700. E = epithelial cell; M = matrigel; L = lumen; P = plastic.
Effect of Matrigel on PSA and PAP Secretion by Isolated Prostatic Epithelial Cells
The secretion of PSA and PAP by epithelial cells grown on plastic or matrigel is shown in Table I. Cells grown on plastic exhibited relatively low levels of PSA secretion which were not significantly altered by the addition of 10 nM dihydrotestosterone (DHT) or 20% stromal cell conditioned medium (SCM). However, the addition of 10 nM DHT and 20% SCM in combination significantly (P < .05) increased PSA secretion by 1.6-fold and 0.9-fold compared to media control or 20% SCM plus DHT, respectively. Cultivation of epithelial cells on matrigel resulted in a greater than 8-fold increase in PSA secretion compared to cells grown on plastic which was further enhanced by the addition of either DHT or SCM. The combination of DHT and SCM stimulated higher levels of PSA secretion than either agent alone although the increase was not statistically significant compared to SCM alone. In all cases, the levels of PSA observed in the presence of 20% SCM were significantly (P < .05) greater than with 20% unconditioned medium. Assessment of PAP secretion showed relatively low levels of PAP in conditioned medium obtained from epithelial cells cultured on plastic. PAP secretion was significantly enhanced by the addition of 10 nM DHT and 20% SCM in combination but not by each agent alone. Cultivation of epithelial cells on matrigel increased PAP secretion by greater than 5-fold compared to plastic. As with PSA, the addition of DHT and SCM, either alone or in combination, resulted in further enhancement of PAP secretion.
228
Fong et al.
Fig. 3. Electron micrographs of epithelial cells cultured on matrigel. A: Development of lumen-like structure following cultivation of epithelial cells on reconstituted basement membrane for 12 days. L = lumen. X9,IOO. B: Presence of tight junctions and associated cytoskeletal elements after 5 days of culture. Arrows demonstrate tight junctions. X 10,OOO. C: Presence of numerous secretory vesicles in epithelial cells cultured on matrigel for 12 days. Arrows designate secretory vesicles. E = epithelial cell nucleus; M = matrigel. x 7.100.
The presence of PSA and PAP in epithelial cell conditioned medium after cultivation on matrigel was confirmed by immunoprecipitation and two-dimensional electrophoresis as shown in Figure 5. Immunoblotting of prostatic fluid using antibodies to PSA and PAP demonstrated the presence of PSA and PAP in their respective 32 kD and 45 kD positions (Fig. 5A). Immunoprecipitation of epithelial cell conditioned medium obtained after cultivation of cells on matrigel with antibodies to PSA and PAP also demonstrated the presence of these major prostatic secretory products (Fig. 5B,C). PSA and PAP were not detected in conditioned medium obtained from epithelial cells cultured on plastic using this technique (data not shown).
Human Prostate Epithelial Cell Biology
229
Plastic
6 4
Mat rigel
I
01 0
I
I
I
I
1
2
3
4
I
I
I
5
6
7
Days Fig. 4. Growth curves of epithelial cells grown on plastic or basement membrane. Cells (5 x IOJ/well) were plated in plastic or matrigel-coated 24 well plates and cultured for 6 days. Cells were harvested using dispase and viable cells were counted using a hemacytometer on days 2, 4, and 6.
TABLE I. Secretion of PSA and PAP by Prostatic Epithelial Cells Cultured on Plastic or Reconstituted Basement Membranet PAP (ng/105 cells) Group Control 10 nM DHT 20% SCM 10 nM DHT + 20% SCM 20% control medium 10 nM DHT + 20% control medium
Plastic
Matrigel
1 . 1 f 0.1
5.8 t 0.3* 10.9 ? 1.3*+** 10.0 k 1 . 5 * 3 * * 15.5 k 1.4**** 8.6 ? l . I * *
1.4 2 0.2 1.1 2 0.2 2.5 ? 0.1* 1.0 2 0.2 1.3 2 0.1
9.4
?
0.7*.**
PSA (ng/105 cells) Plastic Matrigel 1.5 f 0.2 1.6 2 0.1 1.8 2 0.4 3.9 ? 0.4* 1.4 f 0.1
14.6 f 0.2** 24.5 t 2.5**** 28.9 t 4.4**** 33.1 t 2.2*9** 18.2 t 1.8**
1.6
21.7 t 1.3****
?
0.1
'Epithelial cells were cultured on plastic or matrigel for 10 days and conditioned medium was evaluated for PSA and PAP content by radioimmunoassay. Cell counts were performed using a hemacytometer. Control cells were cultured in complete WAJC 404 medium. Experimental cultures were treated with 10 nM DHT and/or 20% prostatic stromal cell conditioned medium. Twenty percent non-conditioned medium was utilized as an additional control. *P < .05 compared to control. **P < .05 compared to cells cultured on plastic.
Intermediate Filament Expression in Prostatic Epithelial Cells
The effect of the tissue culture environment on cytokeratin expression by prostatic epithelial cells is demonstrated in Figure 6. Ananlysis of fresh prostatic tissue using anti-cytokeratin 8.12 showed expression of cytokeratin polypeptides 15 and 19
230
Fong et al.
Fig. 5. lmmunoprecipitation of PSA and PAP from epithelial cell conditioned medium. PSA and PAP were immunoprecipitated from epithelial cell conditioned medium using polyclonal antibodies. Immunoprecipitated proteins were separated using two-dimensional electrophoresis. For comparison, the electrophoretic mobility of PSA and PAP in expressed prostatic fluid was demonstrated by western immunoblotting. A Immunoblotting of PSA and PAP from expressed prostatic fluid. B: Immunoprecipitation of PSA from epithelial cell conditioned medium. C: Immunoprecipitation of PAP from epithelial cell conditioned medium. IgG designates the inimunoprecipitating antibody.
in the human prostatic epithelium according to the convention of Moll and colleagues [ 151. Cytokeratin polypeptides 5, 8, and 18 were observed following immunoblotting with anti-CK 8.13. Epithelial cells isolated directly from percoll gradients exhibited the same cytokeratin profile as fresh prostate. However, cultivation of epithelial cells on either plastic or matrigel resulted in marked alterations in cytokeratin expression. Specifically, proteins exhibiting immunoreactivity with anti-cytokeratins were observed which possessed electrophoretic mobilities identical with cytokeratins 6, 7, 13, 14, 16, and 17 which were not observed in fresh human prostate or epithelial cells harvested from percoll gradients. Assessment of vimentin expression in cultured epithelial cells is shown in Figure 7. Cells cultured on plastic or type IV collagen exhibited expression of the
Human Prostate Epithelial Cell Biology 8.12
231
8.13
Fig. 6. Western blot analysis of cytokeratin expression in cultured epithelial cells and fresh prostatic tissue. Proteins were extracted from fresh prostatic tissue and isolated prostatic epithelial cells and separated using two-dimensional electrophoresis. Cytokeratin expression was determined by western blotting with cytokeratin-specific monoclonal antibodies.
mesenchymal intermediate filament polypeptide vimentin. However, cultivation of epithelial cells on laminin-coated surfaces resulted in suppression of vimentin expression. DISCUSSION Recent advances in tissue culture methodology have allowed for the in vitro propogation of freshly isolated benign epithelial cells from the human prostate [ 16,171. These advancements have enhanced the potential for the effective study of prostate cell biology. Current culture systems utilize serum-free media which provide a sub-physiological extracellular calcium concentration [ 16-1 81. These media preparations allow for the rapid proliferation of benign prostatic epithelial cells, an important factor in obtaining sufficient numbers of epithelial cells for study. The goal of the present study was to evaluate the ability of cultured prostatic epithelial cells to express characteristics which are normally associated with the human prostate in vivo. Results of our studies show that cultivation of prostatic epithelial cells on standard tissue culture plastic results in rapid monolayer growth and also a loss of differentiation-related properties such as secretion of PSA and PAP. In vivo, the
232
Fong et al.
Fig. 7 . Immunocytochemical analysis of vimentin expression in epithelial cells cultured on plastic, type IV collagen, or laminin. Epithelial cells were cultured for 5 days on plastic, type IV collagen, or laminin and fixed with 100% methanol. The presence of vimentin was determined by immunoperoxidase staining using monoclonal anti-vimentin. A: Plastic. B: Type IV collagen. C: Laminin. ~ 4 0 0 .
basolateral surface of the prostatic epithelium rests in intimate contact with basement membrane. We utilized reconstituted basement membrane as a culture substrate and observed enhanced secretion of PSA and PAP as well as promotion of differentiated morphology. Cultivation of prostatic epithelial cells on reconstituted basement membrane resulted in markedly reduced cellular proliferation compared to cells grown on plastic. In vivo, the normal prostatic epithelium maintains a critical balance between cell death and proliferation which results in the maintenance of epithelial homeostasis [19]. Even in the case of benign prostatic hyperplasia, the expansion of prostatic epithelial and stromal elements occurs over a period of many years [20]. Therefore, the study of prostatic epithelial cell growth under conditions that allow for rapid proliferation may not produce optimal results. Indeed, studies have shown that the response of corneal and mammary epithelial cells to growth factors can vary greatly depending on the culture surface utilized [2 1,221. Therefore, considerations should be given to the growth surface used when studying the response of prostatic epithelial cells to putative growth regulators. Our studies show that the reduced growth rate observed following cultivation of epithelial cells on matrigel is associated with greatly enhanced secretion of PSA and PAP, the major secretory products of the intact prostatic epithelium. Previous studies have shown that cultivation of mouse mammary epithelial cells or colon carcinoma cells on reconstituted basement membrane will maintain enhanced differentiated function and morphological features typical of a differentiated epithelium [23-261. Herein, we utilized reconstituted basement membrane as a culture substrate and observed the induction of differentiated morphology and enhanced secretory activity in primary human prostatic epithelial cell cultures. Our morphological studies indicate that cultivation of prostatic epithelial cells on matrigel results in the maintenance of several properties which are associated with the prostatic epithelium in vivo such as the formation of lumen-like structures, tight junctions with associated cytoskeletal elements, and abundant secretory vesicles. Collectively, these findings indicate that many properties of a differentiated epithelium can be maintained by cultivation of prostatic epithelial cells on reconstituted basement membrane. In further studies, we assessed the effect of DHT and prostatic stromal cell
Human Prostate Epithelial Cell Biology
233
conditioned medium on the secretory activity of cultured epithelial cells. The effect of androgens on prostate epithelial cell function has been well established [27,28]. Studies have also suggested that secretory products from stromal cells can influence the growth and secretory activity of prostatic epithelial cells [2,3]. Our studies show that both DHT and SCM were effective in promoting PSA and PAP secretion by prostatic epithelial cells and that these agents were more effective in combination than when added alone. The chemical and functional identity of stromal cell-produced growth and differentiation factors remains to be defined. The benign prostatic epithelium possesses a specific pattern of intermediate filament expression [13,14]. Specifically, cytokeratin polypeptides 5, 8, 15, 18, and 19 have been observed in fresh specimens of normal and hyperplastic human prostate. The cytokeratins are a class of intermediate filaments which are found in epithelial and mesothelial cells [15,29]. Each epithelial type exhibits a specific pattern of cytokeratin expression which can be examined as a marker of epithelial cell differentiation. In the present study, we observed marked alterations in cytokeratin expression in cultured prostatic epithelial cells. The observed changes were present in cells cultured on plastic or matrigel. These findings indicate that, although cultivation of epithelial cells on basement membrane can promote some features of epithelial cell differentiation, cytokeratin expression remained abnormal. An additional member of the intermediate filament family is vimentin. Vimentin is a cytoskeletal polypeptide that is normally expressed in mesenchymal and mesothelial cells but not in epithelia [30]. Several studies have shown that in vitro propogation of epithelial cells, including those from the prostate, will induce expression of vimentin [5]. Our studies show that cultivation of prostatic epithelial cells on the basement membrane component laminin, but not type IV collagen, will suprress vimentin expression. These findings indicate that some, but not all, aspects of intermediate filament expression can be normalized by cultivation of prostatic epithelial cells on basement membrane components. The regulation of prostatic epithelial cell growth and differentiation is a complex problem that is influenced by local and systemic factors. The development of effective culture techniques for human prostatic epithelial cells has provided an important system for the study of prostatic epithelial cell biology. The present study shows that some, but not all, properties of a differentiated prostatic epithelium can be maintained by culturing prostatic epithelial cells on reconstituted basement membrane using currently defined media preparations. We also show that androgens and stromal secretory products have a significant impact on epithelial function in vitro. Future investigations will further define the interactions of local and systemic factors in the regulation of growth and differentiation in the human prostate. ACKNOWLEDGMENTS
These studies were supported by NIH grand DK 39250 and the Lucy and Edwin Kretschmer Fund of Northwestern University Medical School. REFERENCES I . Lee C, Sensibar J: Proteins of the rat prostate. 11. Synthesis of new proteins in the ventral lobe during castration-inducedregression. J Urol 138:903-908, 1987.
Fong et al.
234
2. Kabalin N, Peehl D, Stamey T: Clonal growth of human prostate epithelial cells is stimulated by fibroblasts. The Prostate 14:251-263, 1989. 3. Djakiew D, Tarkington M, Lynch J: Paracrine stimulation of polarized secretion from monolayers of a neoplastic prostatic epithelial cell line by prostatic stromal cell proteins. Cancer Res 50: 1966-1974, 1990.
4. Kyprianou N, lsaacs J: Expression of transforming growth factor beta in the rat ventral prostate during castration-induced programmed cell death. Mol Endocrinol 3:1515-1522, 1989. 5. Shenvood E, Berg L, McEwan R, Pasciak R, Kozlowski J, Lee C: Two-dimensional protein profiles of cultured epithelial and stromal cells from hyperplastic human prostate. J Cell Biochem 40: 201-214, 1989. 6. Kozlowski J, McEwan R, Keer H, Sensibar J, Sherwood E, Lee C, Grayhack J, Albini A, Martin G: Prostate cancer and the invasive phenotype: Application of new in vivo and in vitro approaches. In Fidler IJ, Nicholson G (eds): “Tumor Growth and Progression.” New York: Alan R. Liss, Inc., 1988, pp 189-231. 7. Kleinman H, McGarvey M, Hassell J , Starr V, Cannon F, Laurie J, Martin G: Basement membrane complexes with biological activity. Biochemistry 25312-318, 1986. 8. Bauer KD, Clevenger CV, Endow RK, Murad T, Epstein .4L, Scarpelli DG: Simultaneous nuclear antigen and DNA content quantitation using paraffin-embedded colonic tissue and multiparameter flow cytometry. Cancer Res 46:2428-2435, 1986. 9. Dean P, Jett J: Mathematical analysis of DNA distributions derived from flow microfluorometry. J Cell Biol 60523-527. 1974. 10. Anderson NG, Anderson NL: Analytic techniques for cell fractions: XXI. Two-dimensional analysis of serum and tissue proteins: Multiple isoelectric focusing. Anal Biochem 85:331-340, 1978. I I . Anderson NL, Anderson NG: Analytical techniques for cell fractions: XXL. Two-dimensional analysis of serum and tissue proteins. Multiple gradient slab electrophoresis. Anal Biochem 85:341-352, 1978. 12. Guevera J, Johnston DA, Ranagali LS, Martin BA, Capatillo S, Rodriguez LV: Qualitative aspects of silver deposition in proteins resolved in complex polyacrylamide gels. Electrophoresis 3: 197-206, 1982. 13. Sherwood E, Berg L, Mitchell N , McNeal J, Kozlowski J, Lee C: Differential cytokeratin expression in normal, hyperplastic and malignant epithelial cells from human prostate. J Urol 143:167-171, 1990. 14. Sherwood E, Berg L, Theyer G , Steiner G, Kozlowski J , Lee C: Differential expression of specific cytokeratin polypeptides in the basal and luminal epithelia of the human prostate. The Prostate 18:303-314, 1991. 15. Moll R, Franke W, Schiller D: The catalog of human cytokeratins: Patterns of expression in normal epithelia, tumors and cultured cells. Cell 31:11-25, 1982. 16. Lechner J, Narayan K, Ohnuki Y, Babcock M, Jones L, Kaighn M: Replicative epithelial cell cultures from normal human prostate gland. JNCl 60:797, 1978. 17. Peehl D: Serial culture of adult human prostatic epithelial cells. J Tissue Culture Methods 953-60, 1985. 18. McKeehan WL, Adams PS, Rosser MP: Direct mitogenic effects of insulin, epidermal growth factors, glucocorticoid, cholera toxin, unknown pituitary factors and possibly prolactin, but not androgen, on normal rat prostate epithelial cells in serum-free, primary cell culture. Cancer Res 44:1998-206, 1984. 19. Lee C, Sensibar J , Dudek S, Hiipakka R, Liao S: Prostate ductal system in rats. Regional variation in morphological and functional activities. Biol Reprod 43: 1079-1086, 1990. 20. Grayhack JT, Kozlowski JM: Benign prostatic hyperplasia. In Gillenwater J, Grayhack JT, Howards SS, Duckett JW (eds): “Adult and Pediatric Urology.” Chicago: Yearbook Publishers, 1988, pp 1062-1 125. 21. Miller F, McEachern D, Miller B: Growth regulation of mouse mammary tumor cells in collagen gel cultures by diffusible factors produced by mammary gland epithelium and stromal fibroblasts. Cancer Res 496091-6097, 1989. 22. Gospodarwicz D, Greenburg G , Birdwell C: Determination of cellular shape by the extracellular matrix and its correlation with the control of cellular growth. Cancer Res 38:4155-4171, 1978. 23. Li ML, Judith A, Deborah AF, Carroll H, John H, Bissell MJ: Influence of a reconstituted basement
Human Prostate Epithelial Cell Biology
24. 25. 26. 27. 28.
29. 30.
235
membrane and its components on casein gene expression and secretion in mouse mammary epithelial cells. Roc Natl Acad Sci USA 84:136-140, 1987. Lee E, Lee W, Katezel C, Parry G, Bissell M: Interaction of mouse mammary epithelial cells with collagen substrata: Regulation of casein gene expression and secretion. Proc Natl Acad Sci USA 82: 1419-1423, 1985. Barcellos-Hoff M, Aggeler J, Ram T, Bissell M: Functional differentiation and alveolar morphogenesis of primary mammary cultures on reconstituted basement membrane. Development 105: 223-235, 1989. Daneker G, Piazza A, Steele G, Mercurio A: Relationship between extracellular matrix interactions and degree of differentiation in human colon carcinoma cell lines. Cancer Res 49:681-686, 1989. Huggins C, Hodges C: Studies on prostatic cancer. I. The effect of castration, of estrogen and of androgen injection on serum phosphatases in metastatic carcinoma of the prostate. Cancer Res 1~293-297, 1941. Grayhack J, Keeler T, Kozlowski J: Carcinoma of the prostate. Hormonal therapy. Cancer 60: 589-601, 1987. Wu Y,Parker L, Binder N, Beckert M, Senard 3, Griffiths C, Rheinwald J: The mesothelial keratins: A new family of cytoskeletal proteins identified in cultured mesothelial cells and non-keratinizing epithelia. Cell 31:693-698, 1982. Gown A, Vogel A: Monoclonal antibodies to intermediate filament proteins of human cells: Unique and cross-reacting antibodies. J Cell Biol 91:414-424, 1982.
Reconstituted Basement Membrane Promotes Morphological and Functional Differentiation of Primary Human Prostatic Epithelial Cells Chau-Jye Fong, Edward R. Sherwood, Debra M. Sutkowski, Graziella M. Abu-Jawdeh, Hidejiro Yokoo, Kenneth D. Bauer, James M. Kozlowski, and Chung Lee Departments of Urology (C.-J.F., E.R.S., D.M.S., J.M.K., C.L.), and Pathology (G.M.A.-J., H. Y., K.D.B.), Northwestern University Medical School, Chicago, Illinois Prostatic epithelial cells undergo rapid proliferation and lose their ability to synthesize and secrete prostate-specific antigen (PSA) and prostatic acid phosphatase (PAP) under standard tissue culture conditions. Herein, we compared the morphology, growth, secretory activity, and intermediate filament expression of human prostatic epithelial cells cultured on either standard tissue culture plastic or reconstituted basement membrane. Epithelial cells grown on plastic exhibited a 10-fold increase in proliferation and a higher percentage of cells in the S-phase of the cell cycle compared to cells cultured on basement membrane. However, cells grown on basement membrane secreted markedly higher levels of PSA and PAP. The basement membrane-induced enhancement of secretory activity was potentiated by dihydrotestosterone (DHT) and prostate stromal cell conditioned medium. Morphological studies showed that cells plated on basement membrane formed organoid-like clusters and maintained several aspects of differentiated epithelium including abundant secretory vesicles, microvilli, and desmosomes with associated cytoskeletal elements. Cultivation of epithelial cells on basement membrane components also suppressed the expression of vimentin, a mesenchymal intermediate filament polypeptide. However, cytokeratin expression was abnormal in cells grown on either surface. These results indicate that the differentiated properties of prostatic epithelial cells are promoted by cultivation on reconstituted basement membrane in the presence of DHT and stromal cell conditioned medium.
Key words: PSA, PAP, intermediate filaments
INTRODUCTION The mature secretory epithelium of the human prostate consists of highly differentiated cells that secrete a complex prostatic fluid. In vivo, the prostatic epithelium resides in an environment in which factors such as androgens, extracellular matrix, and stroma influence cellular growth and differentiation [ 1-41. In conventional culture, prostatic epithelial cells undergo a few rounds of cell division and rapidly lose differentiated function. Thus, their characteristics in vitro may have little relevance to
Received for publication April 2, 1991; accepted June 10, 1991 Address reprint requests to Chau-lye Fong, Department of Urology, Northwestern University Medical School, 303 East Chicago Avenue, Chicago, IL 60611.
0 1991 Wiley-Liss, Inc.
222
Fong et al.
the in vivo condition. This observation is further reflected in the inability of prostatic epithelial cells to synthesize and secrete prostatic acid phosphatase (PAP) and prostate-specific antigen (PSA) after cultivation on tissue culture plastic [5-61. Therefore, in order to effectively study prostatic epithelial cell biology, there is a need to establish culture conditions which maintain the normal functional characteristics of the prostatic epithelium. In previous studies, we reported techniques for the isolation and selective cultivation of human prostatic epithelial cells [5-6). In the present study, we attempted to define the role of basement membrane in the regulation of prostatic epithelial cell function. Specifically, we studied the cellular proliferation, morphology, secretory activity, and intermediate filament expression of prostatic epithelial cells cultured on either plastic or reconstituted basement membrane. We also assessed the impact of androgen and stromal cell conditioned medium on the secretory activity of primary prostatic epithelial cell cultures. MATERIALS AND METHODS Isolation of Epithelial and Stromal Cells From the Human Prostate
Prostatic tissue was obtained from individuals undergoing suprapubic or retropubic prostatectomy for treatment of bladder neck obstruction secondary to benign prostatic hyperplasia (BPH). Representative portims of the specimens were submitted for histopathologic evaluation to assure the diagnosis of BPH without focal prostatic carcinoma. The remaining tissue was minced into 1 mm3 fragments and transferred to sterile dissociation flasks containing RPMI- 1640 medium supplemented with 10%fetal bovine serum, 200 U/ml of collagenase type I (Sigma Chemical, St. Louis), and 100 pg/ml of DNAse type I (Sigma). The tissue was dissociated for 16-18 hours at 37°C using a magnetic stimng bar to provide gentle agitation. Following the dissociation period, the resulting cell suspension was washed, resuspended in Hanks’ balanced salt solution (HBSS), and layered over discontinuous percoll (Pharmacia, Uppsala, Sweden) gradients for separation as previously described [5-61. Epithelial cell aggregates were aspirated from gradients, washed (2 x ) with HBSS, and utilized in our studies. In all studies, cells were plated using WAJC 404 medium supplemented with 10% fetal bovine serum (plating medium). Twentyfour hours after plating, the media were changed into serum-free WAJC 404 supplemented with ITS (insulin at 5 pg/ml, transfemn at 5 pg/ml, selenous acid at 5 ng/ml, Collaborative Research, Bedford, MA) epidermal growth factor (3 ng/ml, Sigma), bovine pituitary extract (30 pg/ml, Collaborative Research), prolactin (3 ng/ml, Sigma), cholera toxin (10 ng/ml, Sigma), polyvinyl pyrrolidone (2 mg/ml, Berhring Diagnostics, La Jolla, CA), and penicillin (100 U/ml)/streptomycin (100 pg/ml). Stromal cells were harvested from gradients, washed, and cultured in RPMI- 1640 medium containing 10%fetal bovine serum. Stromal cell conditioned medium was prepared by culturing confluent monolayers in RPMI-I640 containing ITS for 48 hours. Assessment of Prostatic Epithelial Cell Morphology
Isolated epithelial cells were harvested directly from percoll gradients and plated (2 X 105/well) onto untreated 24 well plates (Falcon-Becton Dickinson, Lincoln Park, NJ) or 24 well plates coated with reconstituted basement membrane (Ma-
Human Prostate Epithelial Cell Biology
223
trigel, Collaborative Research, Bedford, MA). Matrigel consists of laminin, type IV collagen, heparan sulfate proteoglycans, and small amounts of entactin and nidogen [7]. The following day (day 0), the media were changed into serum-free WAJC 404 medium. Cells were cultured for 6 days and light photomicroscopy was performed periodically using a phase contrast microscope to document morphology. In electron microscopy studies, cells were cultured on plastic or matrigel as outlined above. After 5-12 days of cultivation, cells were washed (3 X ) with HBSS and fixed with Millonig’s fixative for 30 minutes at 4°C. Cells were washed (3 X ) with HBSS, dehydrated by successive incubations with 25-100% ethanol, and embedded in Epon. Thin sections were cut perpendicular to the dish surface and doubly stained with lead citrate and uranyl acetate for analysis by transmission electron microscopy. Mitogenesis Assay
Prostatic epithelial cells (5 X lo4 cells/well) were plated onto plastic or matrigel-coated surfaces in 24 well plates as outlined above and allowed to adhere overnight (16-18 hours). The following day (day 0), the cultures were washed (3 X ) with HBSS and day 0 cell counts were performed to determine plating efficiency in each group. Cells were cultured for 6 days in serum-free WAJC 404 and counted on days 2, 4, and 6 to assess growth. Existing media were replaced with fresh media on days 2 and 4. Cellular viability was determined using 0.25% trypan blue after detaching the cells from the culture surfaces. In all experiments, cells were detached from matrigel-coated surfaces using dispase (Collaborative Research) and from plastic using 0.1% trypsin (Hazleton, Lenexa, KS). Viable cells were counted using a hemacytometer. In all studies, cellular viability was greater than 90%.
Flow Cytometry DNA content and the cell cycle distribution of prostatic epithelial cells grown on plastic or matrigel for 6 days were determined using flow cytometry. The cells were plated in low density (2 X 104/well),so that after 6 days of culture the cells were less than 80% confluent. The cells were harvested, washed (3 X ) with HBSS, and resuspended (2 x 106/ml)in RPMI 1640 medium supplemented with 10% FBS and 10% DMSO. Cellular DNA was stained with propidium iodide using the Triton-X method [8]. DNA content was analyzed by measuring red fluorescence on the an Epics 752 flow cytometer (Coulter Corp, Hialeah, FL). Cell cycle analysis was performed using Multicycle software (Phoenix Flow Systems, San Diego, CA) which uses the algorithm of Dean and Jett [9]. Diploid chronic lymphocytic leukemia (CLL) cells were utilized as controls for DNA content determinations. Determination of Prostate Specific Antigen (PSA) and Prostatic Acid Phosphatase (PAP) Secretion by Prostatic Epithelial Cells Prostatic epithelial cells (8 X lo5 cells/well) were plated in 24 well plates and allowed to adhere overnight. The following day, media were changed to serum-free WAJC 404 containing specific additives and cells were cultured for 10 days. Media were replaced every other day with fresh media containing appropriate additives. Conditioned media were harvested from each well on day 10, centrifuged (400g for 10 minutes) to remove residual cells, and stored at -80°C. The cells in each well
224
Fong et al.
were harvested and counted on a hemacytometer. PSA and PAP levels in conditioned media were determined using the Tandem-R PAP and PSA immunoradiometric assay (Hybritech, San Diego, CA). For PAP, 100 pl of conditioned medium was added to 12 x 75 mm plastic tubes followed by monoclonal anti-PAP IgG-coated beads and 100 p1of '251-labeled anti-PAP IgG tracer antibody. The samples were incubated for 4 hours, washed (2 X ) with phosphate-buffered saline (PBS) and counted in an LKB minigamma system. PSA was analyzed in an identical fashion except that 50 p1 of conditioned medium was used and monoclonal antibodies were directed against specific epitopes on the PSA molecule. The radioactivity was directly proportional to the concentration of PAP or PSA in the test sample, which was determined from a standard curve.
lmmunoprecipitation Prostatic epithelial cells (5 X 10' cells/well) were plated in 24 well plates using WAJC 404 with 10% FBS and allowed to adhere overnight. The following day, the media were changed into serum-free WAJC 404 and cells were cultured for 8 days with media being changed every other day. The conditioned media were harvested and centrifuged (400g for 10 minutes) to remove residual cells. PAP and PSA were immunoprecipitated from epithelial cell conditioned media using Protein A sepharose CL-6B beads (Pharmacia) which were washed (2 X ) with phosphate-buffered saline (PBS) and resuspended at a concentration of 100 mg/ml in PBS. Twenty microliters of anti-PSA or anti-PAP (Dako, Santa Barbara, CA) was added to the slurry and incubated (4°C) for 6 hours with constant agitation. The beads were washed (3 X ) with PBS to remove excess antibody and resuspended in PBS as above. Five hundred microliters of protein A sepharose slurry was added to 20 ml of prostatic epithelial conditioned medium. Samples were incubated for 16 hours with constant agitation, washed (3 X ) with PBS, and eluted from the beads with urea mix (9 M urea, LKB ampholyte pH 9-1 1, 2% mercaptoethanol, 4% NP-40) for two-dimensional electrophoresis. For comparison, PSA and PAP were identified in expressed prostatic fluid by western immunoblotting as outlined below. Two-Dimensional Electrophoresis The ISO-DALT system for two-dimensional electrophoresis was performed according to the method of Anderson and Anderson [10,11]. Briefly, the first dimension (the IS0 system), which separates proteins according to their isoelectric points, was carried out for 14,000 volt hours in 5% acrylamide containing 8 M urea, 2% total ampholytes with a pH range of 3-10, and 1.7% nonidet detergent. At the end of the isoelectric focusing, each gel was extruded from the elongated glass tube (17.8 x 0.15 cm) and equilibrated in a solution containing 138 mM Tris base, 10% glycerol, 2% sodium dodecyl sulfate (SDS), and bromphenol blue (pH 6.8). The second dimension (the DALT system) separated proteins according to their molecular weights. The slab gels consisted of a linear polyacrylamide gradient [9-I 81 containing 1% SDS. The electrophoresis buffer contained 24 mM Tris, 0.2 M glycine, and 1% SDS. Electrophoresis was carried out at a constant voltage of 100-150 volts for 16 hours. Protein in gels was detected by the silver staining method of Guevara and colleagues [ 121. Western immunoblotting was performed by transferring proteins to nitrocellulose (Bio-Rad, 0.45 p.m) at 120 volts for 2 hours using 25 mM Tris, 192
Human Prostate Epithelial Cell Biology
225
mM glycine, and 20% methanol transfer buffer. Blots were washed (3 x ) with distilled water, incubated with blocking buffer (Carnation non-fat dry milk) for 2 hours and with primary antibody overnight. The blots were washed (3 X ) with blocking buffer and incubated with horseradish peroxidase-conjugated anti-mouse IgG for 2 hours. The color reaction was induced using 4-chloro- l-naphthol substrate. Analysis of Intermediate Filament Expression
For analysis of cytokeratin expression, epithelial cells were cultured on plastic or matrigel for 6 days. Cells were harvested and washed (3 X ) with PBS. Cells were dissolved in urea mix and proteins were separated using two-dimensional electrophoresis. Cytokeratins were analyzed by western immunoblotting using monoclonal anti-cytokeratins 8.12 and 8.13 (Sigma Chemical). For comparison, the cytokeratin profiles of fresh prostatic tissue and epithelial cells isolated directly from percoll gradients were determined as previously described [ 13,141. Vimentin gene expression was assessed by immunocytochemical analysis of purified epithelial cell cultures grown on plastic or surfaces coated with the basement membrane components type IV collagen or laminin. Type IV collagen and laminin (Collaborative Research) -coated surfaces were prepared by addition of 0.5 ml of collagen or laminin solution (10 kg/ml) to 24 well plates. The wells were allowed to air dry under aseptic conditions, and epithelial cells ( 5 X 104/well) were plated onto respective wells and cultured for 3 days. Cells were fixed with 100% methanol, washed (3 x ) with PBS, and incubated (4°C) overnight with mouse anti-human vimentin (Sigma Chemical Co., Clone VIM 13.2). Cells were washed (3 X ) with PBS, and immunoreactivity was determined using the Vectastain (Vector Laboratories, Burlingame, CA) ABC immunoperoxidase system and diaminobenzidine as substrate.
RESULTS Morphology of Prostatic Epithelial Cells Grown on Plastic or Matrigel Cultivation of human prostatic epithelial cells on matrigel was associated with the adoption of a clustered morphology that was not expressed when cells were maintained on plastic as shown in Figure 1. Specifically, cells cultured on plastic exhibited monolayer growth and fusiform morphology. Epithelial cells cultured on plastic formed subconfluent monolayers after 2 days of cultivation which grew to confluence by day 6. In contrast, cells cultured on matrigel for 2 days exhibited a rounded morphology. Between days 2 and 6, cells which were initially randomly dispersed over the growth surface migrated to form organoid-like clusters. By day 6, well-formed clusters were completed and could be maintained for more than 3 weeks. Electron microscopic analysis of epithelial cells cultured on plastic or matrigel is shown in Figure 2. Analysis of cell clusters obtained following cultivation of epithelial cells on matrigel for 5 days showed the presence of several layers of epithelial cells which possessed numerous microvilli (Fig. 2A). Cells grown on matrigel for 12 days also exhibited abundant microvilli as well as numerous secretory vesicles and luminal structures containing secretory material (Fig. 2B). Cells cultured on plastic grew in monolayer and possessed few microvilli or secretory vesicles (Fig. 2C). The presence of differentiation-associated structures in epithelial cells cultured
226
Fong et al. PLASTIC
MATRIGEL
Fig. I . Phase contrast micrographs of prostatic epithelial cells cultured on plastic or matrigel. Cells were plated onto plastic or matrigel-coated surfaces and cultured for 6 days. Cells were photographed on days 2 and 6 to demonstrate cellular morphology. Arrows show clusters of epithelial cells observed following cultivation of cells on matrigel for 6 days. ~ 4 0 0 .
on matrigel is further demonstrated in Figure 3. The presence of luminal structures with associated microvilli was observed following cultivation of epithelial cells on matrigel for 12 days (Fig. 3A). Epithelial cells cultured on matrigel for 5 days also possessed desmosomes with associated cytoskeletal elements (Fig. 3B). Desmosomes were not observed in epithelial cells cultured on plastic. Numerous secretory vesicles and pseudopodial extensions into the reconstituted basement membrane were also observed following 12 days of culture on matrigel (Fig. 3C). Growth of Prostatic Epithelial Cells on Plastic or Matrigel The proliferation of prostatic epithelial cells on plastic and matrigel was determined using cell counts as outlined in Figure 4. Cells cultured on plastic exhibited a doubling time of 42 hours during the log phase of growth and showed a greater than 5-fold increase in cell number over the 6 day cultivation period. Cells cultured on matrigel did not exhibit significant proliferation during the 6 days of cultivation but maintained a constant population of viable cells during the culture period. How cytometry studies confirmed the enhanced growth of cells cultured on plastic. Cells grown on plastic had a higher number of cells in the S phase of the cell cycle compared to cells grown on basement membrane matrigel (15.7% versus 9.3%). Both cell populations exhibited a diploid DNA content as determined by comparison to diploid chronic lymphocytic leukemia cells.
Human Prostate Epithelial Cell Biology
227
Fig. 2. Electron micrographs of epithelial cells cultured on plastic or matrigel. Cells were cultured on plastic or matrigel-coated surfaces and prepared for electron microscopy. A: Epithelial cells cultured on matrigel for 5 days, X 2,000. B: Epithelial cells cultured on matrigel for 12 days, X 4,000. C: Epithelial cells cultured on plastic for 5 days, X 2,700. E = epithelial cell; M = matrigel; L = lumen; P = plastic.
Effect of Matrigel on PSA and PAP Secretion by Isolated Prostatic Epithelial Cells
The secretion of PSA and PAP by epithelial cells grown on plastic or matrigel is shown in Table I. Cells grown on plastic exhibited relatively low levels of PSA secretion which were not significantly altered by the addition of 10 nM dihydrotestosterone (DHT) or 20% stromal cell conditioned medium (SCM). However, the addition of 10 nM DHT and 20% SCM in combination significantly (P < .05) increased PSA secretion by 1.6-fold and 0.9-fold compared to media control or 20% SCM plus DHT, respectively. Cultivation of epithelial cells on matrigel resulted in a greater than 8-fold increase in PSA secretion compared to cells grown on plastic which was further enhanced by the addition of either DHT or SCM. The combination of DHT and SCM stimulated higher levels of PSA secretion than either agent alone although the increase was not statistically significant compared to SCM alone. In all cases, the levels of PSA observed in the presence of 20% SCM were significantly (P < .05) greater than with 20% unconditioned medium. Assessment of PAP secretion showed relatively low levels of PAP in conditioned medium obtained from epithelial cells cultured on plastic. PAP secretion was significantly enhanced by the addition of 10 nM DHT and 20% SCM in combination but not by each agent alone. Cultivation of epithelial cells on matrigel increased PAP secretion by greater than 5-fold compared to plastic. As with PSA, the addition of DHT and SCM, either alone or in combination, resulted in further enhancement of PAP secretion.
228
Fong et al.
Fig. 3. Electron micrographs of epithelial cells cultured on matrigel. A: Development of lumen-like structure following cultivation of epithelial cells on reconstituted basement membrane for 12 days. L = lumen. X9,IOO. B: Presence of tight junctions and associated cytoskeletal elements after 5 days of culture. Arrows demonstrate tight junctions. X 10,OOO. C: Presence of numerous secretory vesicles in epithelial cells cultured on matrigel for 12 days. Arrows designate secretory vesicles. E = epithelial cell nucleus; M = matrigel. x 7.100.
The presence of PSA and PAP in epithelial cell conditioned medium after cultivation on matrigel was confirmed by immunoprecipitation and two-dimensional electrophoresis as shown in Figure 5. Immunoblotting of prostatic fluid using antibodies to PSA and PAP demonstrated the presence of PSA and PAP in their respective 32 kD and 45 kD positions (Fig. 5A). Immunoprecipitation of epithelial cell conditioned medium obtained after cultivation of cells on matrigel with antibodies to PSA and PAP also demonstrated the presence of these major prostatic secretory products (Fig. 5B,C). PSA and PAP were not detected in conditioned medium obtained from epithelial cells cultured on plastic using this technique (data not shown).
Human Prostate Epithelial Cell Biology
229
Plastic
6 4
Mat rigel
I
01 0
I
I
I
I
1
2
3
4
I
I
I
5
6
7
Days Fig. 4. Growth curves of epithelial cells grown on plastic or basement membrane. Cells (5 x IOJ/well) were plated in plastic or matrigel-coated 24 well plates and cultured for 6 days. Cells were harvested using dispase and viable cells were counted using a hemacytometer on days 2, 4, and 6.
TABLE I. Secretion of PSA and PAP by Prostatic Epithelial Cells Cultured on Plastic or Reconstituted Basement Membranet PAP (ng/105 cells) Group Control 10 nM DHT 20% SCM 10 nM DHT + 20% SCM 20% control medium 10 nM DHT + 20% control medium
Plastic
Matrigel
1 . 1 f 0.1
5.8 t 0.3* 10.9 ? 1.3*+** 10.0 k 1 . 5 * 3 * * 15.5 k 1.4**** 8.6 ? l . I * *
1.4 2 0.2 1.1 2 0.2 2.5 ? 0.1* 1.0 2 0.2 1.3 2 0.1
9.4
?
0.7*.**
PSA (ng/105 cells) Plastic Matrigel 1.5 f 0.2 1.6 2 0.1 1.8 2 0.4 3.9 ? 0.4* 1.4 f 0.1
14.6 f 0.2** 24.5 t 2.5**** 28.9 t 4.4**** 33.1 t 2.2*9** 18.2 t 1.8**
1.6
21.7 t 1.3****
?
0.1
'Epithelial cells were cultured on plastic or matrigel for 10 days and conditioned medium was evaluated for PSA and PAP content by radioimmunoassay. Cell counts were performed using a hemacytometer. Control cells were cultured in complete WAJC 404 medium. Experimental cultures were treated with 10 nM DHT and/or 20% prostatic stromal cell conditioned medium. Twenty percent non-conditioned medium was utilized as an additional control. *P < .05 compared to control. **P < .05 compared to cells cultured on plastic.
Intermediate Filament Expression in Prostatic Epithelial Cells
The effect of the tissue culture environment on cytokeratin expression by prostatic epithelial cells is demonstrated in Figure 6. Ananlysis of fresh prostatic tissue using anti-cytokeratin 8.12 showed expression of cytokeratin polypeptides 15 and 19
230
Fong et al.
Fig. 5. lmmunoprecipitation of PSA and PAP from epithelial cell conditioned medium. PSA and PAP were immunoprecipitated from epithelial cell conditioned medium using polyclonal antibodies. Immunoprecipitated proteins were separated using two-dimensional electrophoresis. For comparison, the electrophoretic mobility of PSA and PAP in expressed prostatic fluid was demonstrated by western immunoblotting. A Immunoblotting of PSA and PAP from expressed prostatic fluid. B: Immunoprecipitation of PSA from epithelial cell conditioned medium. C: Immunoprecipitation of PAP from epithelial cell conditioned medium. IgG designates the inimunoprecipitating antibody.
in the human prostatic epithelium according to the convention of Moll and colleagues [ 151. Cytokeratin polypeptides 5, 8, and 18 were observed following immunoblotting with anti-CK 8.13. Epithelial cells isolated directly from percoll gradients exhibited the same cytokeratin profile as fresh prostate. However, cultivation of epithelial cells on either plastic or matrigel resulted in marked alterations in cytokeratin expression. Specifically, proteins exhibiting immunoreactivity with anti-cytokeratins were observed which possessed electrophoretic mobilities identical with cytokeratins 6, 7, 13, 14, 16, and 17 which were not observed in fresh human prostate or epithelial cells harvested from percoll gradients. Assessment of vimentin expression in cultured epithelial cells is shown in Figure 7. Cells cultured on plastic or type IV collagen exhibited expression of the
Human Prostate Epithelial Cell Biology 8.12
231
8.13
Fig. 6. Western blot analysis of cytokeratin expression in cultured epithelial cells and fresh prostatic tissue. Proteins were extracted from fresh prostatic tissue and isolated prostatic epithelial cells and separated using two-dimensional electrophoresis. Cytokeratin expression was determined by western blotting with cytokeratin-specific monoclonal antibodies.
mesenchymal intermediate filament polypeptide vimentin. However, cultivation of epithelial cells on laminin-coated surfaces resulted in suppression of vimentin expression. DISCUSSION Recent advances in tissue culture methodology have allowed for the in vitro propogation of freshly isolated benign epithelial cells from the human prostate [ 16,171. These advancements have enhanced the potential for the effective study of prostate cell biology. Current culture systems utilize serum-free media which provide a sub-physiological extracellular calcium concentration [ 16-1 81. These media preparations allow for the rapid proliferation of benign prostatic epithelial cells, an important factor in obtaining sufficient numbers of epithelial cells for study. The goal of the present study was to evaluate the ability of cultured prostatic epithelial cells to express characteristics which are normally associated with the human prostate in vivo. Results of our studies show that cultivation of prostatic epithelial cells on standard tissue culture plastic results in rapid monolayer growth and also a loss of differentiation-related properties such as secretion of PSA and PAP. In vivo, the
232
Fong et al.
Fig. 7 . Immunocytochemical analysis of vimentin expression in epithelial cells cultured on plastic, type IV collagen, or laminin. Epithelial cells were cultured for 5 days on plastic, type IV collagen, or laminin and fixed with 100% methanol. The presence of vimentin was determined by immunoperoxidase staining using monoclonal anti-vimentin. A: Plastic. B: Type IV collagen. C: Laminin. ~ 4 0 0 .
basolateral surface of the prostatic epithelium rests in intimate contact with basement membrane. We utilized reconstituted basement membrane as a culture substrate and observed enhanced secretion of PSA and PAP as well as promotion of differentiated morphology. Cultivation of prostatic epithelial cells on reconstituted basement membrane resulted in markedly reduced cellular proliferation compared to cells grown on plastic. In vivo, the normal prostatic epithelium maintains a critical balance between cell death and proliferation which results in the maintenance of epithelial homeostasis [19]. Even in the case of benign prostatic hyperplasia, the expansion of prostatic epithelial and stromal elements occurs over a period of many years [20]. Therefore, the study of prostatic epithelial cell growth under conditions that allow for rapid proliferation may not produce optimal results. Indeed, studies have shown that the response of corneal and mammary epithelial cells to growth factors can vary greatly depending on the culture surface utilized [2 1,221. Therefore, considerations should be given to the growth surface used when studying the response of prostatic epithelial cells to putative growth regulators. Our studies show that the reduced growth rate observed following cultivation of epithelial cells on matrigel is associated with greatly enhanced secretion of PSA and PAP, the major secretory products of the intact prostatic epithelium. Previous studies have shown that cultivation of mouse mammary epithelial cells or colon carcinoma cells on reconstituted basement membrane will maintain enhanced differentiated function and morphological features typical of a differentiated epithelium [23-261. Herein, we utilized reconstituted basement membrane as a culture substrate and observed the induction of differentiated morphology and enhanced secretory activity in primary human prostatic epithelial cell cultures. Our morphological studies indicate that cultivation of prostatic epithelial cells on matrigel results in the maintenance of several properties which are associated with the prostatic epithelium in vivo such as the formation of lumen-like structures, tight junctions with associated cytoskeletal elements, and abundant secretory vesicles. Collectively, these findings indicate that many properties of a differentiated epithelium can be maintained by cultivation of prostatic epithelial cells on reconstituted basement membrane. In further studies, we assessed the effect of DHT and prostatic stromal cell
Human Prostate Epithelial Cell Biology
233
conditioned medium on the secretory activity of cultured epithelial cells. The effect of androgens on prostate epithelial cell function has been well established [27,28]. Studies have also suggested that secretory products from stromal cells can influence the growth and secretory activity of prostatic epithelial cells [2,3]. Our studies show that both DHT and SCM were effective in promoting PSA and PAP secretion by prostatic epithelial cells and that these agents were more effective in combination than when added alone. The chemical and functional identity of stromal cell-produced growth and differentiation factors remains to be defined. The benign prostatic epithelium possesses a specific pattern of intermediate filament expression [13,14]. Specifically, cytokeratin polypeptides 5, 8, 15, 18, and 19 have been observed in fresh specimens of normal and hyperplastic human prostate. The cytokeratins are a class of intermediate filaments which are found in epithelial and mesothelial cells [15,29]. Each epithelial type exhibits a specific pattern of cytokeratin expression which can be examined as a marker of epithelial cell differentiation. In the present study, we observed marked alterations in cytokeratin expression in cultured prostatic epithelial cells. The observed changes were present in cells cultured on plastic or matrigel. These findings indicate that, although cultivation of epithelial cells on basement membrane can promote some features of epithelial cell differentiation, cytokeratin expression remained abnormal. An additional member of the intermediate filament family is vimentin. Vimentin is a cytoskeletal polypeptide that is normally expressed in mesenchymal and mesothelial cells but not in epithelia [30]. Several studies have shown that in vitro propogation of epithelial cells, including those from the prostate, will induce expression of vimentin [5]. Our studies show that cultivation of prostatic epithelial cells on the basement membrane component laminin, but not type IV collagen, will suprress vimentin expression. These findings indicate that some, but not all, aspects of intermediate filament expression can be normalized by cultivation of prostatic epithelial cells on basement membrane components. The regulation of prostatic epithelial cell growth and differentiation is a complex problem that is influenced by local and systemic factors. The development of effective culture techniques for human prostatic epithelial cells has provided an important system for the study of prostatic epithelial cell biology. The present study shows that some, but not all, properties of a differentiated prostatic epithelium can be maintained by culturing prostatic epithelial cells on reconstituted basement membrane using currently defined media preparations. We also show that androgens and stromal secretory products have a significant impact on epithelial function in vitro. Future investigations will further define the interactions of local and systemic factors in the regulation of growth and differentiation in the human prostate. ACKNOWLEDGMENTS
These studies were supported by NIH grand DK 39250 and the Lucy and Edwin Kretschmer Fund of Northwestern University Medical School. REFERENCES I . Lee C, Sensibar J: Proteins of the rat prostate. 11. Synthesis of new proteins in the ventral lobe during castration-inducedregression. J Urol 138:903-908, 1987.
Fong et al.
234
2. Kabalin N, Peehl D, Stamey T: Clonal growth of human prostate epithelial cells is stimulated by fibroblasts. The Prostate 14:251-263, 1989. 3. Djakiew D, Tarkington M, Lynch J: Paracrine stimulation of polarized secretion from monolayers of a neoplastic prostatic epithelial cell line by prostatic stromal cell proteins. Cancer Res 50: 1966-1974, 1990.
4. Kyprianou N, lsaacs J: Expression of transforming growth factor beta in the rat ventral prostate during castration-induced programmed cell death. Mol Endocrinol 3:1515-1522, 1989. 5. Shenvood E, Berg L, McEwan R, Pasciak R, Kozlowski J, Lee C: Two-dimensional protein profiles of cultured epithelial and stromal cells from hyperplastic human prostate. J Cell Biochem 40: 201-214, 1989. 6. Kozlowski J, McEwan R, Keer H, Sensibar J, Sherwood E, Lee C, Grayhack J, Albini A, Martin G: Prostate cancer and the invasive phenotype: Application of new in vivo and in vitro approaches. In Fidler IJ, Nicholson G (eds): “Tumor Growth and Progression.” New York: Alan R. Liss, Inc., 1988, pp 189-231. 7. Kleinman H, McGarvey M, Hassell J , Starr V, Cannon F, Laurie J, Martin G: Basement membrane complexes with biological activity. Biochemistry 25312-318, 1986. 8. Bauer KD, Clevenger CV, Endow RK, Murad T, Epstein .4L, Scarpelli DG: Simultaneous nuclear antigen and DNA content quantitation using paraffin-embedded colonic tissue and multiparameter flow cytometry. Cancer Res 46:2428-2435, 1986. 9. Dean P, Jett J: Mathematical analysis of DNA distributions derived from flow microfluorometry. J Cell Biol 60523-527. 1974. 10. Anderson NG, Anderson NL: Analytic techniques for cell fractions: XXI. Two-dimensional analysis of serum and tissue proteins: Multiple isoelectric focusing. Anal Biochem 85:331-340, 1978. I I . Anderson NL, Anderson NG: Analytical techniques for cell fractions: XXL. Two-dimensional analysis of serum and tissue proteins. Multiple gradient slab electrophoresis. Anal Biochem 85:341-352, 1978. 12. Guevera J, Johnston DA, Ranagali LS, Martin BA, Capatillo S, Rodriguez LV: Qualitative aspects of silver deposition in proteins resolved in complex polyacrylamide gels. Electrophoresis 3: 197-206, 1982. 13. Sherwood E, Berg L, Mitchell N , McNeal J, Kozlowski J, Lee C: Differential cytokeratin expression in normal, hyperplastic and malignant epithelial cells from human prostate. J Urol 143:167-171, 1990. 14. Sherwood E, Berg L, Theyer G , Steiner G, Kozlowski J , Lee C: Differential expression of specific cytokeratin polypeptides in the basal and luminal epithelia of the human prostate. The Prostate 18:303-314, 1991. 15. Moll R, Franke W, Schiller D: The catalog of human cytokeratins: Patterns of expression in normal epithelia, tumors and cultured cells. Cell 31:11-25, 1982. 16. Lechner J, Narayan K, Ohnuki Y, Babcock M, Jones L, Kaighn M: Replicative epithelial cell cultures from normal human prostate gland. JNCl 60:797, 1978. 17. Peehl D: Serial culture of adult human prostatic epithelial cells. J Tissue Culture Methods 953-60, 1985. 18. McKeehan WL, Adams PS, Rosser MP: Direct mitogenic effects of insulin, epidermal growth factors, glucocorticoid, cholera toxin, unknown pituitary factors and possibly prolactin, but not androgen, on normal rat prostate epithelial cells in serum-free, primary cell culture. Cancer Res 44:1998-206, 1984. 19. Lee C, Sensibar J , Dudek S, Hiipakka R, Liao S: Prostate ductal system in rats. Regional variation in morphological and functional activities. Biol Reprod 43: 1079-1086, 1990. 20. Grayhack JT, Kozlowski JM: Benign prostatic hyperplasia. In Gillenwater J, Grayhack JT, Howards SS, Duckett JW (eds): “Adult and Pediatric Urology.” Chicago: Yearbook Publishers, 1988, pp 1062-1 125. 21. Miller F, McEachern D, Miller B: Growth regulation of mouse mammary tumor cells in collagen gel cultures by diffusible factors produced by mammary gland epithelium and stromal fibroblasts. Cancer Res 496091-6097, 1989. 22. Gospodarwicz D, Greenburg G , Birdwell C: Determination of cellular shape by the extracellular matrix and its correlation with the control of cellular growth. Cancer Res 38:4155-4171, 1978. 23. Li ML, Judith A, Deborah AF, Carroll H, John H, Bissell MJ: Influence of a reconstituted basement
Human Prostate Epithelial Cell Biology
24. 25. 26. 27. 28.
29. 30.
235
membrane and its components on casein gene expression and secretion in mouse mammary epithelial cells. Roc Natl Acad Sci USA 84:136-140, 1987. Lee E, Lee W, Katezel C, Parry G, Bissell M: Interaction of mouse mammary epithelial cells with collagen substrata: Regulation of casein gene expression and secretion. Proc Natl Acad Sci USA 82: 1419-1423, 1985. Barcellos-Hoff M, Aggeler J, Ram T, Bissell M: Functional differentiation and alveolar morphogenesis of primary mammary cultures on reconstituted basement membrane. Development 105: 223-235, 1989. Daneker G, Piazza A, Steele G, Mercurio A: Relationship between extracellular matrix interactions and degree of differentiation in human colon carcinoma cell lines. Cancer Res 49:681-686, 1989. Huggins C, Hodges C: Studies on prostatic cancer. I. The effect of castration, of estrogen and of androgen injection on serum phosphatases in metastatic carcinoma of the prostate. Cancer Res 1~293-297, 1941. Grayhack J, Keeler T, Kozlowski J: Carcinoma of the prostate. Hormonal therapy. Cancer 60: 589-601, 1987. Wu Y,Parker L, Binder N, Beckert M, Senard 3, Griffiths C, Rheinwald J: The mesothelial keratins: A new family of cytoskeletal proteins identified in cultured mesothelial cells and non-keratinizing epithelia. Cell 31:693-698, 1982. Gown A, Vogel A: Monoclonal antibodies to intermediate filament proteins of human cells: Unique and cross-reacting antibodies. J Cell Biol 91:414-424, 1982.
