0013.7227/92/1305-2955$03.00/O Endocrinology Copyright 0 1992 by The Endocrine

Basic Fibroblast Stromal Growth EDWARD R. SHERWOOD1-, JAMES M. KOZLOWSKI Department

of

Vol. Printed

Society

Urology, Northwestern

Growth Factor: A Potential in the Human Prostate* CHAU-JYE University

FONG, Medical

CHUNG

LEE,

Studies were undertaken, using isolated prostatic epithelial and stromal cells, to evaluate the role of basic fibroblast growth factor (bFGF) in the regulation of benign prostatic growth. bFGF was detected in lysates, but not the conditioned media, of cultured prostatic epithelial and stromal cells by Western immunoblotting and immunoprecipitation of metabolically labeled proteins. Immunofluorescence analysis of benign human prostate localized the majority of bFGF to the prostatic stroma. In addition, bFGF was a potent stimulator of stromal cell proliferation in uitro, but was not mitogenic to cultured epithelial cells. Further studies demonstrated bFGF receptors (Kd = 258 PM; 61,400 receptors/cell) on stromal cells, but not epithelial cells. Epithelial cell-conditioned medium was mitogenic for stromal cells, suggesting the presence of paracrine

Mediator

of

AND

School, Chicago, Illinois

ABSTRACT.

130, No. 5 in U.S.A.

60611

interactions. However, bFGF does not appear to be the mediator of this interaction, since the mitogenic effect of epithelial cellconditioned medium on stromal cells was not significantly reduced by the addition of anti-bFGF. Additional studies showed that concentrated stromal cell-conditioned medium was not mitogenic to cultured stromal cells under serum-free defined conditions, indicating the lack of an external autocine mechanism. These studies demonstrate that bFGF is actively synthesized by isolated prostatic epithelial and stromal cells, but is largely not secreted. Prostatic stroma, but not epithelia, are responsive to the mitogenic effect of bFGF in uitro. However, because of the limited secretion of bFGF by prostatic cells, the mechanism(s) of bFGF-mediated regulation of stromal growth remains unclear. (Endocrinology 130: 2955-2963, 1992)

N

UMEROUS investigators have observed fibroblast growth factors (FGFs) in human and rat prostates (l-6). These investigations have resulted in the identification of fibroblast growth factors in both benign and malignant prostatic tissue. Basic FGF (bFGF) and acidic FGF (aFGF) have received the most intense study, and both have been identified in the rat prostate (6). However, aFGF does not appear to be a local growth regulator in the human prostate. Studies by Mydlo et al. (3) and Mori et al. (7) have shown that bFGF, but not aFGF, is actively synthesized in the human prostate. The role of bFGF in the regulation of human prostatic growth remains largely undefined. The site(s) of bFGF synthesis and the identity of bFGF-responsive cells in the human prostate have not been fully ascertained. Story and colleagues (8) have shown that cultured prostatic stromal cells actively synthesize bFGF and are responsive to the mitogenic effect of bFGF. However, the

ability of stromal cells to actively secrete bFGF has not been determined. Furthermore, there are no reports of bFGF synthesis and secretion or the expression of bFGF receptors by human prostatic epithelial cells. In the present study we used techniques for the selective cultivation of human prostatic epithelial and stromal cells to assessthe synthesis and secretion of bFGF by the respective prostatic cell types. Further studies were undertaken to determine which prostatic cells express bFGF receptors and are responsive to the mitogenic effects of bFGF. We also evaluated the site(s) of bFGF concentration in fresh prostatic tissue. The goal of these studies was to further define the role of bFGF in the local regulation of human prostatic growth. Materials

and Methods

Growth factors and antibodies bFGF and aFGF were purchased from R&D Systems (Minneapolis, MN). Transforming growth factor-a (TGFcu) was purchased from ICN Biochemicals (Costa Mesa, CA). Epidermal growth factor (EGF) and platelet-derived growth factor (PDGF) were purchased from Collaborative Research (Bedford, MA). Rabbit antibovine bFGF was purchased from R&D Systems. Monoclonal mouse anticytokeratin (clone 8.13) and antivimentin (clone VIM-13.2) were purchased from Sigma Chemical Co. (St. Louis, MO). Fluorescein isothiocyanate-con-

Received October 17, 1991. Address all correspondence and requests for reprints to: Dr. Edward R. Sherwood, Department of Urology, Northwestern University Medical School, 303 East Chicago Avenue, Chicago, Illinois 60611. * This work was surmorted in Dart bv NIH Grant DK-39250 and the Lucy and Edwin Kretschmer Fund of Nbrthwestern University Medical School. t Recipient of NIH Postdoctoral Fellowship DK-08204 and the William 0. Jeffery III Fellowship for Prostate Cancer Research of Northwestern University Medical School.

2955

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 November 2015. at 08:44 For personal use only. No other uses without permission. . All rights reserved.

2956

bFGF

AND

HUMAN

jugated rabbit antimouse immunoglobulin G (IgG) and goat antirabbit IgG were purchased from Boehringer-Mannheim (Indianapolis, IN). Radioiodinated bFGF (935 Ci/mM) was purchased from Amersham Corp. (Arlington Heights, IL). Radioiodinated EGF (93.8 &i/pg) was purchased from ICN Radiochemicals (Costa Mesa, CA). Isolation

of prostatic

epithelial

and stromal cells

Epithelial and stromal cells were isolated from prostatic tissue obtained from patients undergoing open prostatectomy to relieve bladder neck obstruction secondary to benign prostatic hyperplasia (BPH). The diagnosis of BPH without focal prostatic carcinoma was confirmed by review of pathology reports and histological evaluation of hematoxylinand eosinstained tissue sections in our laboratory. Purified epithelial and stromal cells were isolated from fresh BPH tissue as previously outlined (9, 10). Briefly, prostatic tissue was dissociated overnight (16-18 h) using type I collagenase (200 U/ml; Sigma Chemical) and DNAse (100 pg/ml; Sigma) in RPMI-1640 medium containing 10% fetal bovine serum. The epithelial and stromal populations were separated using discontinuous Percoll gradient centrifugation. Stromal cells were maintained in continuous culture using RPMI-1640 medium containing 10% fetal bovine serum and penicillin (100 U/ml)-streptomicin (100 pg/ ml). Epithelial cells were cultured in complete WAJC 404 medium containing bovine pituitary extract (30 pg/ml; Collaborative Research), EGF (10 rig/ml; Collaborative Research), cholera toxin (10 rig/ml; Sigma), PRL (3 rig/ml; Sigma), ITS (Collaborative Research), and penicillin-streptomicin. Basal WAJC for mitogenesis assays consisted of all of the above additives with the exception of bovine pituitary extract and EGF. Epithelial cells were used at the first or second serial passage. Stromal cells were harvested for experimentation after three to six serial passages. Our studies show that prostatic epithelial cells synthesize and secrete prostate-specific antigen and prostatic acid phosphatase after isolation with these techniques (11). Immunofluorescence Isolated epithelial and stromal cells were plated onto glass coverslips, cultured for 48 h, and fixed with 100% methanol for 5 min. After washing (three times) with PBS, fixed cells were incubated (37 C) with anticytokeratin or antivimentin (1:20 dilutions) for 1 h. Cells were again washed (three times) with PBS and incubated (three times) with fluorescein-conjugated rabbit antimouse IgG (1:20) for 1 h. Green fluorescence was evaluated using a Leitz Laborlux 11 fluorescence microscope (Rockleigh, NJ). bFGF was localized in normal and hyperplastic prostatic tissue by immunofluorescence staining of frozen sections. Normal prostate was obtained from donors less than 25 yr of age through the cooperation of the Regional Organ Bank of Illinois. Hyperplastic prostate was derived from men undergoing open prostatectomy, as previously described. Methanol-fixed sections were incubated (37 C) with primary polyclonal anti-bFGF (1:250) for 1 h. Control slides were incubated with anti-bFGF that was preincubated (4 C) with 1 pg purified bovine bFGF. After washing, specimens were incubated with fluorescein-

Endo. Voll30.

PROSTATE

1992 No 5

conjugated goat antirabbit IgG (1:20) for 1 h (37 C) and evaluated by fluorescence microscopy. Preparation

of conditioned

media

Epithelial or stromal cells were grown to near confluence in 150-cm2 tissue culture flasks (Corning Glass Works, Corning, NY). Conditioned media were collected by washing cells (three times) with PBS and incubating stromal (RPMI-1640) or epithelial cells (WAJC 404) in medium supplemented with ITS+ (Collaborative Research) for 48 h. The resulting conditioned media were centrifuged (400 X g for 10 min) to remove residual cells and filtered through a.0.45~Km membrane (Nalge Co., Rochester, NY). Conditioned media were dialyzed against distilled water (4 C) using a 3.5kilodalton (kDa) cut-off membrane (Spectrum Laboratories, Los Angeles, CA) and lyophilized. RPMI-1640 with ITS+ or WAJC 404 with ITS+ were incubated under cell-free conditions and served as controls for stroma-conditioned medium and epithelia-conditioned medium, respectively. Mitogenesis

assay

Epithelial or stromal cells (1 X 105/flask) were plated in 25cm2 tissue culture flasks (Corning Glass Works) and allowed to adhere overnight (16-18 h). Cells were washed (three times) with PBS, and experimental samples were added to each flask. Cells were counted with a Coulter counter (Coulter Electronics, Hialeah, FL) after detaching the cells from the culture flask with fresh 0.25% trypsin-0.1% EDTA. The media in the remaining flasks were replaced with fresh media containing appropriate additives at the time of cell counting. Metabolic

labeling

Epithelial and stromal cells were cultured to near confluence, washed (three times) with PBS, and metabolically labeled by incubating cells in cystine/methionine-free medium (Sigma) supplemented with ITS+ and 200 &i each of [““Slmethionine (184 mCi/mM; Amersham) and cysteine (106 mCi/mM; Amersham) for 24 h. Conditioned media were aspirated, and cells were washed (three times) with PBS and dissolved in lysis buffer (0.5% Triton X-100, 300 nM NaCl, 50 mM Tris, 10 mM iodoacetamide, 1 mM phenylmethylsulfonylfluoride, and 10 pg/ ml leupeptin). Cell lysates were cleared by centrifugation (10,000 X g for 10 min). Heparin-affinity

purification

and immunoprecipitation

Heparin-Sepharose CL-6B (Pharmacia, Uppsala, Sweden) was equilibrated with PBS (100 mg beads/ml) and incubated (4 C) with conditioned media or cell lysates (125 ~1 beads/ml sample) for 16-18 h. Beads were washed (three times) with PBS, and proteins were eluted with PBS containing 2.5 M NaCl. The resulting samples were dialyzed against distilled water using a membrane with 3.5-kDa cut-off, lyophilized, and reconstituted in lysis buffer. For immunoprecipitation, antibody-laden protein-A-Sepharose CL-4B beads (Pharmacia, Uppsala, Sweden) were prepared by incubating (4 C) preswelled beads (100 mg/ml in PBS) with antibody for 16-18 h. Beads were washed (three times) with PBS and added to 20 ml conditioned media (500 ~1) or 5 ml cell lysate (100 ~1). After

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 November 2015. at 08:44 For personal use only. No other uses without permission. . All rights reserved.

bFGF

AND

HUMAN

16-18 h of incubation (4 C), the protein-A beads were washed (five times) with PBS, and proteins were eluted with Laemmli buffer [0.5 M Tris-2% sodium dodecyl sulfate (SDS)]. Samples were mixed 1:l with loading buffer (5% SDS, 20% glycerol, 5% 2-mercaptoethanol, and bromphenol blue) and separated by SDS-polyacrylamide gel electrophoresis (SDS-PAGE), as outlined below. Radiolabeled proteins were visualized by fluorography, using Kodak X-Omat film (Rochester, NY).

SDS-PAGE and Westernimmunoblotting Samples, partially purified by heparin binding, were mixed 1:l with loading buffer and loaded onto 12.5% polyacrylamide gels. Gels were run at 60 mamp for 4-5 h using a resolving buffer consisting of 25 mM Tris, 192 mM glycine, and 5% SDS. Resolved proteins were transferred to nitrocellulose (Bio-Rad; 0.22 pm) membranes according to the method of Towbin and colleagues (12). Briefly, proteins were transferred at 250 mamp for 2 h in a transfer buffer consisting of 25 mM Tris, 192 mM glycine, and 20% methanol. Blots were blocked in carnation dry milk for 2 h and incubated (25 C) with primary anti-bFGF for 16-18 h. After washing with blocking buffer, blots were incubated with horseradish-peroxidase-conjugated goat antirabbit IgG (Sigma; 1:lOOO) for 2 h. Blots were washed (three times) with TBS buffer (500 mM Tris, 2 M NaCl, and 0.5% Tween-20, pH 7.5), and color reaction was induced using 4chloro-1-naphthol (3 mg/ml in methanol). Apparent mol wt were determined using Rainbow mol wt markers (Amersham).

Binding studies Stromal or epithelial cells (1 X lO”/well) were plated in 24well plates and allowed to adhere overnight. The following day, cells were washed (three times) with unsupplemented RPMI1640 medium and incubated (4 C for 4 h) with lz51-labeled bFGF (0.3-3000 PM) in the presence or absence of a loo-fold excess of cold bFGF excess. Cells were washed (three times) with PBS containing 250 pg/ml heparin, followed by the addition of 2% Triton X-100 in water. Samples were counted in an LKB Minigamma system (Rockville, MD).

Statistics Statistical analyses were performed using analysis of variance and Student’s t test. P < 0.05 was considered statistically significant.

Results Characterization of cultured prostatic epithelial and stromal cells The purity of epithelial and stromal cell cultures was assessedby light microscopy and immunofluorescence staining of cytokeratin and vimentin, as shown in Fig. 1. Giemsa-stained epithelial cells (Fig. 1A) exhibited the typical cobblestone morphology and stained strongly after immunostaining with anticytokeratin (Fig. 1C). The cytokeratin polypeptides form a class of intermediate filaments that are specific for epithelial and mesothelial cells (13, 14). Cultured epithelial cells also exhib-

PROSTATE

2957

ited positive immunoreactivity with antivimentin (Fig. 1E). Vimentin is an intermediate filament polypeptide that is specific for mesenchymal cells in situ, but has been frequently observed in cultured epithelial cells (10, 15). Cultured prostatic stromal cells exhibited an elongated spindle-shaped morphology that is typical of cultured fibroblasts (Fig: 1B). Stromal cells did not exhibit immunoreactivity with anticytokeratin (Fig. lD), but stained strongly after immunostaining with antivimentin (Fig. 1F). Identification cells

of bFGF in cultured epithelial and stromal

The presence of bFGF in cultured epithelial and stromal cells was initially demonstrated by Western blotting with anti-bFGF, as shown in Fig. 2. Immunoblot analysis of cultured epithelial cells showed the presence of an immunoreactive band that comigrated with purified bFGF in epithelial cell lysates, but not in conditioned medium. In these experiments, epithelial cells were cultured in media that were free of bovine pituitary extract for 48 h before the addition of growth factor-free medium for the collection of conditioned media and cell lysates. Bovine pituitary extract is a rich source of some FGFs (16). Analysis of stromal cell lysates also showed the presence of an anti-bFGF immunoreactive protein with an electrophoretic mobility identical to that of the bFGF standard. As with epithelial cells, immunoreactive bFGF was not observed in stromal cell-conditioned medium (SCM). The active synthesis of bFGF by cultured epithelial and stromal cells was evaluated by immunoprecipitation of metabolically labeled proteins, as shown in Fig. 3. Stromal cell lysates exhibited an immunoprecipitable band that comigrated with radioiodinated bFGF on SDSPAGE. An immunoprecipitable band with an electrophoretie mobility identical to that of purified bFGF was also observed in epithelial cell lysates. Immunoprecipitable bFGF was not observed in conditioned media from stromal or epithelial cell cultures (data not shown). Localization of bFGF in benign human prostate in situ Immunofluorescence staining using anti-bFGF was undertaken to determine the site(s) of bFGF localization in benign human prostate, as shown in Fig. 4. A hematoxylin- and eosin-stained section of benign prostate is shown to demonstrate the histological features of the prostate (Fig. 4A). A prostatic epithelial acinar structure and associated lumen as well as surrounding stroma are demonstrated. The stromal region of benign prostate showed strong immunostaining with anti-bFGF. The epithelium also exhibited immunoreactivity with antibFGF, but to a lesser degree. As a control, anti-bFGF was preabsorbed by incubation with an excess of purified

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 November 2015. at 08:44 For personal use only. No other uses without permission. . All rights reserved.

2958

bFGF

AND

HUMAN

PROSTATE

Endo Vol1.30.

l

1992 No 5

FIG. 1. Characterization of cultured epithelial and stromal cells from human prostate. Epithelial and stromal cells were plated onto glass coverslips and cultured for 48 h. Cells were stained with Giemsa or antibodies to intermediate filament polypeptides, as described in Muterials and Methods. A, Epithelial cells stained with Giemsa (x400); B, stromal cells stained with Giemsa (x400); C, epithelial cells immunostained with anticytokeratin 8.13 (x1000); D, stromal cells immunostained with anticytokeratin 8.13 (x1000); E, epithelial cells immunostained with antivimentin (X1000); F, stromal cells immunostained with antivimentin (X1000).

46

FIG. 2. Western immunoblot analysis of bFGF in cultured prostatic cells. Cell lysates and conditioned media from cultured prostatic epithelial and stromal cells, partially purified by heparin affinity, were resolved by SDS-PAGE. Resolved proteins were transferred to nitrocellulose membranes and probed with anti-bFGF. Lane 1, bFGF-positive control (500 ng); lane 2, epithelial cell lysate; lane 3, ECM; lane 4, stromal cell lysate; lane 5, SCM. The arrow indicates bFGF standard.

bFGF. Immunostaining was not observed in control specimens. Several specimens of normal and hyperplastic prostate were evaluated using immunocytochemistry, and all exhibited the staining pattern presented in Fig. 4.

FIG. 3. Immunoprecipitation of bFGF from cultured epithelial and stromal cells. Metabolically labeled proteins were purified by heparin affinity and immunoprecipitated using anti-bFGF, as described in Materials and Methods. Proteins were resolved by 12.5% SDS-PAGE and visualized using fluorography. Lane 1, Radioiodinated EGF; lane 2, radioiodinated bFGF, lane 3, stromal cell lysate; lane 4, epithelial cell lysate. The arrow indicates bFGF standard.

Specific binding of bFGF to stromal and epithelial cells

The specific binding of radioiodinated bFGF to prostatic cells is shown in Fig. 5. Stromal cells exhibited saturable (13.8 fmol/104 cells) binding of radioiodinated bFGF at a ligand concentration of 3 nM (Fig. 5A). Saturable binding of bFGF was not observed in cultured epithelial cells at the same ligand concentrations (Fig.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 November 2015. at 08:44 For personal use only. No other uses without permission. . All rights reserved.

bFGF

AND

HUMAN

PROSTATE

2959

A 15

=m d 8 lo8 T z E b

5A). Scatchard analysis of bFGF binding to stromal cells demonstrated the presence of 61.4 X lo3 high affinity (Kd = 258 PM) bFGF receptors/cell (Fig. 5B). Effect of purified growth factors on the proliferation of cultured epithelial and stromal cells from human prostate

The growth response of isolated epithelial cells to purified bFGF as well as other defined growth factors was analyzed by growth curve analysis, as shown in Table 1. Cells were cultured in either nonsupplemented basal WAJC 404 or WAJC 404 supplemented with various defined growth factors. Complete WAJC 404 served as a positive control. bFGF did not significantly alter epithelial cell proliferation after 8 days of cultivation compared to basal WAJC 404. However, aFGF stimulated a significant 48% increase in epithelial cell number by day 8. TGFa! induced significant 1.6- and 5.9-fold increases in epithelial cell growth after 5 and 8 days of cultivation, respectively. EGF stimulated epithelial cell proliferation on days 5 and 8 by 1.7- and 5.4-fold, respectively. PDGF did not stimulate epithelial cell proliferation at any of the time points examined. The response of epithelial cells to growth factors other than bFGF was evaluated for comparison and to assure that epithelial cells were responsive to mitogens under the experimental conditions used in our studies. The responses of cultured prostatic stromal cells to bFGF and other defined growth factors are shown in

Stroma

/O

.o 0

/

5-/

j

FIG. 4. Immunofluorescence analysis of bFGF in human prostatic tissue. Frozen sections of benign human prostate were immunostained using rabbit antibovine bFGF and fluorescein-conjugated goat antirabbit IgG. As a control, anti-bFGF was preincubated with an excess of bFGF before addition to tissues. A, Hematoxylin and eosin-stained section of prostate (S, stroma; L, lumen of prostatic acinus). B, Normal prostate immunostained with anti-bFGF. C, Control specimen in which anti-bFGF was preincubated with an excess of bFGF. Arrows designate areas of prostatic stroma. All sections were photographed at x400.

r

imVer 0

, 10

20

30

bFGF1

B

Eyithel;. 40

, 50

60

(w/ml)

0.06

6 1,404 receptors/cell s

0.03 0.02 0.01 0.00 \ 0

2

4

6

8

10

12

14

‘,bFGF bound (phi) FIG. 5. Specific binding and Scatchard analysis of bFGF receptors on cultured epithelial and stromal cells from human prostate. Isolated epithelial and stromal cells were incubated (4 Cl in the presence of radioiodinated bFGF (0.3-3000 PM) for 4 h. Specific binding was determined in the presence of a loo-fold cold bFGF excess. A, Specific binding of bFGF to cultured epithelial and stromal cells. B, Scatchard analysis of bFGF binding to cultured stromal cells.

Table 2. Cells were cultured in the presence or absence of defined growth factors, using RPMI-1640 medium supplemented with ITS+. RPMI-1640 supplemented with 10% fetal bovine serum served as a positive control. bFGF was a potent stimulator of stromal cell proliferation, as indicated by significant 1.2-, 3.2-, and 4.8-fold increases in stromal cell growth after 2, 4, and 6 days of cultivation, respectively. aFGF enhanced stromal cell growth by l.l- and 1.8-fold on days 4 and 6, respectively. PDGF also significantly stimulated stromal cell proliferation on days 4 and 6 by 0.9- and 1.5-fold, respectively. TGFcx or EGF increased stromal cell proliferation by 0.7and l.l-fold, respectively, by day 6 of cultivation.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 November 2015. at 08:44 For personal use only. No other uses without permission. . All rights reserved.

bFGF

2960 1. Effects of several defined of isolated prostatic epithelial cells

growth

TABLE

Days Growth

factors

in culture

AND

HUMAN

on the proliferation

(relative

proliferation)

factor 2

Complete WAJC 404 Basal WAJC 404 bFGF (10 rig/ml) aFGF (10 rig/ml) TGFti (10 rig/ml) EGF (10 rig/ml) PDGF (10 rig/ml)

94.4 f 100 84.3 f 119.6 f 85.0 f 73.2 ?I 64.6 f

5 15.7 14.2 4.1 5.5 8.7 3.9

8

329.9 f 25.5” 100 96.3 + 20.6 144.1 + 7.1” 255.1 + 43.9” 272.9 + 36.4 88.5 + 29.9

1290.0 123.7 148.4 688.2 640.9 127.4

+ 192.4” 100 + 21.5 + 6.8” + 49.5” + 65.6” + 20.5

Prostatic epithelial cells (1 x lO’/flask) were plated in 25-cm* tissue culture flasks and allowed to adhere for 16-18 h. Media were changed to complete WAJC 404 (positive control), basal WAJC 404 (negative control), or basal WAJC 404 plus the designated growth factors. Cells were counted on days 2,5, and 8. Media were replaced with fresh media containing appropriate additives on days 2 and 5. Values are expressed as the mean + SE. n = G/group. “Significantly (P < 0.05) greater than basal WAJC 404 negative control. 2. Growth responses fined growth factors

TABLE

of isolated

Days Growth

prostatic

in culture

stromal

(relative

cells to de-

growth)

factor 2

10% FBS ITS+ bFGF (10 rig/ml) aFGF (10 rig/ml) PDGF (10 rig/ml) TGFa (10 rig/ml) EGF (10 rig/ml)

242.7 + 7.5” 100 221.6 f 23.4” 142.3 f 7.4” 133.8 f 28.6 109.7 ? 23.6 78.5 f 4.3

4 698.2 f 46.1” 100 424.4 & 65.9” 206.2 + 14.3” 190.3 + 29.0” 125.3 + 8.9 198.6 + 47.9”

6 876.2 576.2 283.3 250.4 170.6 207.5

+ 47.1” 100 f 31.9” f 21.9 + 33.3” + 24.4” + 18.1”

Stromal cells (1 x 105/flask) were plated in 25-cm* flasks and allowed to adhere for 16-18 h. Media were then changed to RPMI-1640 containing 10% fetal bovine serum (positive control), ITS+ (negative control), or ITS+ plus growth factors. Cells were counted on days 2,4, and 6. The media in remaining flasks were replaced with fresh media containing appropriate additives on days 2 and 4. n = G/group. ’ Significantly (P < 0.05) greater than ITS+ negative control.

Effect of conditioned medium from epithelial and stromal cultures on stromal cell proliferation

The response of stromal cells to concentrated conditioned medium from epithelial and stromal cell cultures was assessed by growth curve analysis, as shown in Table 3. Cultivation of stromal cells with 1 mg/ml epithelial cell-conditioned medium (ECM) significantly increased stromal cell proliferation compared to that of stromal cells cultured in ITS+ or 1 mg/ml reconstituted WAJC 404 medium (negative controls). Addition of anti-bFGF IgG (5 pg/well) to ECM did not significantly diminish the mitogenic effect of ECM on stromal cells compared to that on cells cultured in ECM alone, indicating that bFGF is not likely to be the active mitogen. As a control, anti-bFGF (5 pg/ml) inhibited bFGF-induced proliferation of prostatic stromal cells by 45% after 6 days of cultivation, indicating the ability of anti-bFGF to neu-

PROSTATE

Endo. Vol130.No5

tralize the mitogenic effect of bFGF. Compared to the control, cultivation of isolated stromal cells in stromaconditioned medium (SCM) did not result in enhanced stromal cell proliferation, suggesting the absence of an external autocrine growth loop. Additional studies were undertaken to determine the effect of SCM on epithelial cell proliferation, as shown in Table 4. SCM at 10-1000 pg/ml significantly stimulated epithelial cell proliferation compared to the RPMI1640 control. These findings indicate the presence of a currently undefined stroma-derived paracrine growth mediator that induces epithelial cell proliferation. Discussion

Several investigators have isolated FGFs from extracts of rat and human prostatic tissue (l-6). More recent studies have identified bFGF as the primary FGF in the human prostate (3, 7). However, the site(s) of bFGF production in the human prostate and the role of bFGF in the regulation of epithelial and stromal cell growth have not been determined. The present studies were undertaken to define the cellular source(s) of bFGF and identify bFGF-responsive cells in benign human prostate. Our data show that isolated prostatic epithelial and stromal cells actively synthesize bFGF. In addition, cultured stromal cells, but not epithelial cells, possess bFGF receptor(s) and are responsive to the mitogenic effects of bFGF. bFGF was detected by Western blotting and immunoprecipitation of metabolically labeled proteins from cultured epithelial and stromal cells. These findings indicate that bFGF is actively synthesized by both cell types in uitro. The presence of bFGF mRNA and protein in human prostate has been previously reported (1, 3, 4, 7). Story and co-workers (8) have also demonstrated the active synthesis of bFGF by cultured human prostatic stromal cells. The present study extends these observations by showing that both isolated human prostatic epithelial and stromal cells actively synthesize bFGF. In addition, bFGF was observed in lysates of cultured prostatic cells, but not in conditioned medium. The lack of bFGF in conditioned media is probably due to the absence of a signal peptide on intracellular bFGF that is essential for protein packaging and secretion (16-18). This finding has prompted investigators to postulate that bFGF is secreted by atypical mechanisms or is released only after cell death or damage (19). Therefore, bFGF may play a role in the restructuring and maintainence of prostatic architecture after injury or cell death. Several investigators have described programmed cell death in the prostatic epithelium during castration-induced regression (20,21). Recent studies by Lee and colleagues (22,23) have demonstrated programmed cell death in the

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 November 2015. at 08:44 For personal use only. No other uses without permission. . All rights reserved.

bFGF AND HUMAN TABLE

3. Effect

of conditioned

media

from

stromal

and epithelial

PROSTATE

cell cultures

on stromal Days

2961 cell proliferation

in culture

(relative

growth)

Group 4

2 10% FBS ITS+ ECM (1 mg/ml) ECM (0.1 mg/ml) WAJC 404 (1 mg/ml) WAJC 404 (0.1 mg/ml) ECM (1 mg/ml) + anti-bFGF bFGF (10 rig/ml) bFGF (10 rig/ml) + anti-bFGF SCM (1 mg/ml) SCM (0.1 mg/ml)

155.7 f 100 115.6 + 143.7 + 127.5 f 27.0 f 107.2 + 115.2 f 102.3 f 109.7 + 116.5 f

IgG (5 fig/ml) (5 pg/ml)

6

424.0 f 110.9” 100 159.5 + 3.2” 105.9 * 16.8 96.8 + 5.9 78.6 + 5.9 113.6 + 14.1 217.9 2 17.3” 131.6 -+ 14.2 137.2 + 17.4 128.8 + 19.0

7.2” 22.7 7.1 14.4 19.7 4.2 6.7 6.9 25.2 24.3

584.6 + 59.5” 100 181.9 + 28.2” 82.3 + 3.1 70.5 + 13.7 75.3 + 16.3 152.7 + 14.1” 323.6 + 19.8” 177.4 + 19.8” 110.0 + 15.5 100.0 + 10.1

Stromal cells (1 x 105) were plated in 25-cm* tissue culture flasks and allowed to adhere overnight. Media were changed to RPMI-1640 containing 10% fetal bovine serum (positive control), ITS+ (negative control), or ITS+ plus ECM or SCM. Unconditioned WAJC 404 or RPMI1640 served as additional controls for ECM and SCM, respectively. Cells were counted on days 2, 4, and 6. Media were replaced on days 2 and 4 with fresh media containing appropriate additives. n = g-g/group. ’ Significantly (P < 0.05) greater than ITS+ negative control.

TABLE

4. Effect

of SCM

on epithelial

cell proliferation Days

in culture

(relative

growth)

Additive 5

2 Complete WAJC 404 Basal WAJC 404 SCM (1 mg/ml) SCM (0.1 mg/ml) SCM (0.01 mg/ml) RPM1 1640 (1 mg/ml) RPM1 1640 (0.1 mg/ml)

125.0 f 100 108.8 f 116.3 f 112.5 f 101.3 f 108.8 f

15.0 5.0 22.5 16.3 12.7 13.8

491.3 f 36.3” 100 141.3 f 22.5 283.8 + 8.8” 125.0 + 39.1 127.8 f 9.4 101.3 + 25.0

8 1269.9 f 142.5” 100 305.5 +- 31.5” 467.1 + 44.2” 201.4 + 53.4” 117.8 f 39.7 126.0 2 26.0

Epithelial cells (1 x 105/flask) were plated in 25-cm* flasks and allowed to adhere overnight. Media were then changed (positive control), basal WAJC 404 (negative control), or basal WAJC 404 plus stromal cell conditioned medium (SCM). days 2, 5, and 8. Media were replaced on days 2 and 5. a Significantly (P < 0.05) greater than basal WAJC 404 negative control.

proximal epithelium of the intact rat prostate, which appears to be a normal event in the regulation of prostatic homeostasis. Whether bFGF is released by the prostatic epithelium during these events to play a role in the maintainance of prostatic homeostasis remains to be established. Although bFGF is synthesized in both epithelial and stromal cells, our studies indicate that only prostatic stromal cells are responsive to the mitogenic activity of bFGF. This observation was supported by the presence of high affinity bFGF receptors on cultured stromal cells, but not epithelial cells. These findings suggest that bFGF serves as a stroma-specific mitogen in the human prostate. Paradoxically, epithelial cells exhibited a significant mitogenic response to aFGF. This finding is probably due to the interaction of aFGF with the keratinocyte growth factor (KGF) receptor on epithelial cells. Recent reports by Aaronson and colleagues (24, 25) showed that the affinity of aFGF for the KGF receptor is more than lo-fold greater than that of bFGF. An additional report

to complete Cells were

WAJC 404 counted on

by Peehl and Stamey (26) demonstrated that human prostatic epithelial cells were responsive to KGF. This finding suggests the presence of KGF receptors on prostatic epithelial cells, which allows them to respond to aFGF, but not bFGF. Our studies show that conditioned media from prostatic stromal cells are mitogenic for cultured epithelial cells. The ability of SCM to stimulate epithelial cell proliferation has been previously reported (27,28). However, the identity of the stroma-derived mitogen has not been established. The factor is not likely to be bFGF, since bFGF is not mitogenic for prostatic epithelial cells. We also show that conditioned medium from prostatic epithelial cell cultures is mitogenic for stromal cells. At first glance, bFGF seems likely to be the epithelia-derived stromal cell mitogen, because bFGF is mitogenic for stromal cells and is actively synthesized by epithelial cells. However, the mitogenic effect of ECM was not significantly reduced by the addition of anti-bFGF. This may be due to the inability of epithelial cells to secrete

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 November 2015. at 08:44 For personal use only. No other uses without permission. . All rights reserved.

2962

bFGF

AND

HUMAN

bFGF, as demonstrated in our studies. Djakiew and colleagues (28) have shown that stimulation of human prostatic stromal cell growth by prostatic epithelial cells may be mediated by a nerve growth factor-like peptide. The results of our studies show that concentrated SCM is not mitogenic for cultured stromal cells, indicating the absence of an external autocrine mechanism. Story and colleagues (8) have demonstrated that stromal cell lysates are mitogenic for isolated prostatic stromal cells and that the effect is mediated by bFGF. Our data support these findings by demonstrating the active synthesis of bFGF by cultured prostatic stromal cells. However, the lack of bFGF secretion by stromal cells creates uncertainty as to the potential mechanisms of bFGFmediated autocrine growth regulation. The ability of some growth factors to stimulate proliferation through internal autocrine (intercrine) mechanisms has been proposed (29). Growth factors that act through this pathway stimulate growth by interaction with intracellular receptors. However, the ability of bFGF to act through this mechanism has not been established. Immunohistochemical analysis of bFGF in intact human prostate demonstrated stronger immunoreactivity in the stromal compartment of the gland than in the epithelium. The predominance of bFGF in the prostatic stroma in situ may be explained by the binding of bFGF to connective tissue-associated heparan sulfate proteoglycans, which are abundant in the stromal ground substance (30). Previous studies have shown that heparin or heparan sulfate will bind bFGF and protect it from proteolytic degradation and inactivation (31, 32). Heparin has also been shown to potentiate the mitogenic effect of some heparin-binding growth factors (32). The results of recent studies show that a dual receptor system, involving glycosaminoglycans and high affinity bFGF receptors, may be required for bFGF-induced activity (33). Our data suggest that bFGF is sequestered in the stromaassociated connective tissue of the prostate. Consequently, the stromal matrix may serve as a reservoir for the presentation of bFGF to responsive stromal cells. The present study demonstrates the active synthesis of bFGF by human prostatic epithelial and stromal cells in uitro. bFGF was localized primarily to the prostatic stroma in situ and is a potent stimulator of stromal, but not epithelial, cell proliferation in tissue culture. However, because of the limited secretion of bFGF, its role in prostatic stromal-epithelial cell interactions remains undefined. In addition, the mechanisms mediating the regulation of bFGF and bFGF receptor expression in the prostate are not well understood. The prostate is highly dependent on androgens for the maintainence of structural and functional integrity (20, 21). However, the impact of androgens on bFGF and bFGF receptor expression have not been determined.

PROSTATE

Endo. Voll30.

1992 No 5

References 1. Story MT, Sasse J, Jacobs S, Lawson R 1987 Prostatic growth factor: purification and structural relationship to basic fibioblast growth factor. Biochemistry 263843-3849 2. Nishi N, Matuo Y, Kunitomi K, Takenaka I, Usami M, Kotako T, Wada F 1988 Comparative analysis of grwoth factors in normal and pathologic human prostates. Prostate 13:39-48 3. Mydlo JH, Bulbul M, Richon V, Heston W, Fair WR 1988 Heparinbinding growth factor isolated from human prostate extracts. Prostate 12:343-355 4. Mydlo JH, Michaeli J, Heston W, Fair W 1988 Expression of basic fibroblast growth factor mRNA in benign prostatic hyperplasia and prostatic carcinoma. Prostate 13:241-247 W, Adams P 1988 Heparin-binding growth factor/ 5. McKeehan prostatropin attenuates inhibition of rat prostate tumor epithelial cell growth by transforming growth factor beta. In Vitro Cell Dev Biol 24:243-246 6. Mansson P, Adams P, Ken M, McKeehan W 1989 Heparin binding growth factor gene expression and receptor characteristics in normal rat prostate and two transplantable rat prostate tumors. Cancer Res 49:2485-2494 7. Mori H, Maki M, Oishi K, Jaye M, Igarashi K, Yoshida 0, Hatanaka M 1990 Increased expression of genes for basic fibroblast growth factor and transforming growth factor type beta 2 in human benign prostatic hyperplasia. Prostate 16:71-80 B, Baeten L, Swartz S, Jacobs S, Begun F, 8. Story M, Livingston Lawson R 1989 Cultured human prostate-derived fibroblasts produce a factor that stimulates their growth with properties indistinguishable from basic fibroblast growth factor. Prostate 15:355-365 J, McEwan R, Keer H, Sensibar J, Sherwood E, Lee C, 9. Kozlowski Grayhack J, Albini A, Martin G 1988 Prostate cancer and the invasive phenotype: application of new in vitro and in uiuo approaches. In: Fidler IJ, Nicholson G (eds) Tumor Progression and Metastasis. Liss, New York, pp 189-231 10. Sherwood ER, Berg L, McEwan R, Pasciak R, Kozlowski J, Lee C 1989 Two-dimensional protein profiles of cultured epithelial and stromal cells from hyperplastic human prostate. J Cell Biochem 40~201-214 E, Sutkowski D, Abu-Jawdeh G, Yokoo H, 11. Fong C, Sherwood Bauer K, Kozlowski J, Lee C 1991 Reconstituted basement membrane promotes morphological and functional differentiation of primary human prostatic epithelial cells. Prostate 19:221-235 H, Steahelin T, Gordon J 1979 Electrophoretic transfer of 12. Towbin proteins from polyacrylamide gels to nitrocellulose sheets. Proc Nat1 Acad Sci USA 76:4350-4354 W, Schiller D, Geiger B, Krepler R 1982 The 13. Moll R, Franke catalog of human cytokeratins: patterns of expression in normal eoithelia. tumors and cultured cells. Cell 31.1 l-34 14. Sun T, Tseng S, Huang A, Cooper D, Schermer A, Lynch M, Weiss R, Eichner R 1986 Monoclonal antibody studies of mammalian epithelial keratins: a review. Ann NY Acad Sci 455:307-329 15. Gown A, Vogel A 1982 Monoclonal antibodies to intermediate filament proteins of human cells: Unique and cross-reacting antibodies. J Cell Biol 91:414-424 16. Kurokawa T, Sasada R, Iwane M, Igarashi K 1987 Cloning and expression of cDNA encoding for basic fibroblast nrowth factor. FEBS Lett 213:189-194 17. Abraham JA, Mergia A, Whang J, Tumolo A, Friedman J, Hjerrild K, Gospodarowicz D, Fiddes J 1986 Nucleotide sequence of a bovine clone encoding the angiogenic protein basic fibroblast growth factor. Science 233:545-548 JA, Whang J, Tumolo A, Mergia A, Friedman J, Gospo18. Abraham darowicz D, Fiddes J 1986 Human basic fibroblast growth factor: nucleotide sequence and genomic organization. EMBO J 5:25232528 19. Muthukrishnan L, Warder E, McNeil P 1991 Basic fibroblast growth factor is efficiently released from a cytosolic storage site through plasma membrane disruptions of endothelial cells. J Cell Physiol 148:1-16 20. Lee C, Sensibar J 1987 Proteins of the rat prostate. II. Synthesis of new proteins in the ventral lobe during castration-induced

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 November 2015. at 08:44 For personal use only. No other uses without permission. . All rights reserved.

bFGF

AND

HUMAN

regression. J Urol 138903-908 21. Kyprianou N, Isaacs J 1988 Activation of programmed cell death in the rat ventral prostate after castration. Endocrinology 122:552562 22. Lee C, Sensibar J, Dudek S, Hiipakka R, Liao S 1990 Prostatic ductal system in rats: regional variation in morphological and functional activities. Biol Reprod 43:1079-1086 23. Sensibar J, Griswold M, Sylvester S, Buttyan R, Bardin C, Cheng C, Dudek S, Lee C 1991 Prostatic ductal system in rats: regional variation in localization of an androgen-repressed gene product, sulfated glycoprotein-2. Endocrinology 1282091-2102 24. Miki T, Fleming T, Bottaro D, Rubin J, Ron D, Aaronson S 1991 Expression cDNA cloning of the KGF receptor by creation of a transforming autocrine loop. Science 251:72-75 25. Bottaro D, Rubin J, Ron D, Finch P, Florio C, Aaronson S 1990 Characterization of the receptor for keratinocyte growth factor. J Biol Chem 265:12767-12770 26. Peehl D, Stamey T 1991 Fibroblast growth factors can replace epidermal growth factor for clonal proliferation of human prostatic epithelial cells. J Urol 145:331A

PROSTATE

2963

27. Kabalin N, Peehl D, Stamey T 1989 Clonal growth of human prostate epithelial cells is stimulated by fibroblasts. Prostate 14:251-263 28. Djakiew D, Delsite R, Pflug B, Wrathall J, Lynch J, Onoda M 1991 Regulation of growth by a nerve growth factor-like protein which modulates paracrine interactions between neoplastic epithelial cell line and stromal cells of the human prostate. Cancer Res 51:33043310 29. Browder T. Dunbar C, Neinhuis A 1989 Private and public autocrine loops’in neoplastic cells. Cancer Cells 1:9-17 _ 30. Hascall V, Kimura J 1982 Proteoglycans: isolation and characterization. Methods Enzymol82:769-800 31. Saksela 0. Moscatelli D. Sommer A. Rifkin D 1988 Endothelial cell-derived heparan sulfate binds basic fibroblast growth factor and protects it from proteolytic degradation. J Cell Biol 107:743751 32. Gospodarowicz D, Cheng J 1986 Heparin protects basic and acidic libroblast growth factor from inactivation. J Cell Physiol 128:475484 33. Klagsbrun M, Baird A 1991 A dual receptor system is required for basic fibroblast growth factor activity. Cell 67:229-231

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 November 2015. at 08:44 For personal use only. No other uses without permission. . All rights reserved.

Basic fibroblast growth factor: a potential mediator of stromal growth in the human prostate.

Studies were undertaken, using isolated prostatic epithelial and stromal cells, to evaluate the role of basic fibroblast growth factor (bFGF) in the r...
1MB Sizes 0 Downloads 0 Views