0013-7227/90/1262-0818$02.00/0 Endocrinology Copyright © 1990 by The Endocrine Society
Vol. 126, No. 2 Printed in U.S.A.
Rat Prostate Cancer Cells Contain Functional Receptors for Transforming Growth Factor-/?* SYDNEY A. SHAIN, ALAN L. LIN, JUDITH D. ROGER, AND ANDREW G. KARAGANIS Department of Cellular and Molecular Biology, Southwest Foundation for Biomedical Research, San Antonio, Texas 78284
ABSTRACT. Clonally derived C-, D-, and T-family AXC/SSh
of function is comparable to that of other mammalian cells.
rat prostate cancer cell lines contain transforming growth factor|3 (TGF/3) receptors. The content in C3, Dl, Tl, and T5 cells, respectively, was 8,560 ± 1,450,13,160 ± 1,240, 2,425 ± 490, and 10,540 ± 1,025 sites/cell (mean ± SEM). Respective Kd values were 160 ± 48, 200 ± 53, 24 ± 3, and 115 ± 15 pM (mean ± SEM). T l cell TGF/3 receptor site content and Kd differed significantly from those of other prostate cancer cell lines (P < 0.05). TGF/8 is a bifunctional concentration-dependent modulator of T l and T5 cell thymidine incorporation. At low concentrations, thymidine incorporation was inhibited, whereas as the medium TGFjS content was increased, Tl and T5 cell thymidine incorporation was stimulated. The concentrations of TGF/3 causing half-maximum inhibition of T l or T5 cell thymidine incorporation, respectively, were 0.11 and 0.24 pM, whereas the respective TGF/3 concentrations causing half-maximum stimulation of thymidine incorporation were 14.4 and 134 pM. These findings establish that rat prostate cancer cell sensitivity to TGF/3 inhibition of function is at least 2 orders of magnitude greater than that of most other mammalian cells. In contrast, the sensitivity of rat prostate cancer cells to TGF/3 enhancement
TGF/3 inhibited basic fibroblast growth factor (bFGF) stimulation of T l and T5 cell thymidine incorporation. Because the concentration of bFGF required for half-maximum increase of T5 cell thymidine incorporation was independent of medium TGF/3 content, the effect of TGF/3 is distal to the T5 cell bFGF receptor. In contrast, the concentration of bFGF required for half-maximum increase in T l cell thymidine incorporation increased 5-fold as the medium TGF/3 content was increased; suggesting that the effect of TGF/3 in Tl cells is proximal to the Tl cell bFGF receptor. Our studies establish that rat prostate cancer cells contain functional TGF/3 receptors, imply the presence of functional bFGF receptors, and demonstrate that mitogen modulation of prostate cancer cell function is multifactorial. The finding that TGF/3 is a bifunctional effector of prostate cancer cell DNA synthesis provides some insight into the potential complexity of mitogen modulation of prostate cancer cell proliferation. The mechanism by which these mitogens interact is unknown; however, our studies suggest that some interactive effects may be cell line specific. {Endocrinology 126: 818-825, 1990)
T
HE roles of androgens as principal effectors of prostate differentiation (1) and modulators of normal (1) or neoplastic cell function (2) are well established. However, results of recent studies suggest that growth factors also may importantly affect normal and neoplastic prostate cell function. By example, in vitro studies show that EGF1 and insulin enhance the proliferation of normal rat ventral prostate epithelium (3) and
Received August 21, 1989. Address all correspondence and requests for reprints to: Dr.. Sydney A. Shain, Department of Obstetrics and Gynecology, University of Texas Health Science Center, 7703 Floyd Curl Drive, San Antonio, Texas 78284. * This work was supported in part by Grant DK-39766 from the DHHS. 1 The following abbreviations are used: EGF, epidermal growth factor; TGFa, transforming growth factor-a; TGF/3, transforming growth factor-/3; TGF/31, homodimeric transforming growth factor-/31; aFGF, acidic fibroblast growth factor; bFGF, basic fibroblast growth factor; ECGF, endothelial cell growth factor; 5aDHT, 5a-dihydrotestosterone; FBS, fetal bovine serum; IGF-I, insulin-like growth factorI; PDGF, platelet-derived growth factor; MEMEM, Modified Eagle's Minimum Essential Medium.
that EGF and TGFa enhance the proliferation of normal canine prostate epithelium (4). Rat ventral prostate contains EGF (5) and TGF/3 (6) receptors, and cell receptor content is androgen modulated. Human normal (7), hypertrophic (7), and carcinomatous prostate cells (7, 8) also contain EGF receptors. An EGF-like mitogen (9), bFGF (9), and a 25-kDa acid-stable mitogen lacking transforming activity (10) are produced by normal rat ventral prostate. Human prostate carcinoma PC-3 cells elaborate TGF/3 (11), an amino-terminal extended bFGF has been purified from human benign hyperplastic tissue (12, 13), and bFGF transcripts are detectable in specimens of human prostatic carcinoma (14). Dunning rat prostate carcinoma elaborates a heparin-binding mitogen with properties characteristic of ECGF (15), and we have shown that AXC/SSh rat prostate carcinoma cells secrete mitogens with properties characteristic of the FGF family of growth factors (16). These findings suggest potential interactive roles for androgens and growth factors as modulators of normal and neoplastic prostate cell
818 The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 19:32 For personal use only. No other uses without permission. . All rights reserved.
PROSTATE CANCER CELL TGF0 RECEPTORS function. We have described clonally derived AXC/SSh rat prostate cancer cells which are distinguished as C-, D-, or Tfamily members based on their maintenance on medium containing charcoal-stripped serum which, respectively, has no added steroid or contains either 10~7 M 5«DHT or 10~7 M testosterone (17, 18). We have shown that in vitro proliferation of C- and D-family member cell lines is not androgen modulated (18), that androgens enhance T l cell and diminish T5 cell proliferation in vitro (18), and that the response of AXC/SSh prostate cancer cells to secreted or prototypic mitogens differs (16). Androgen modulation of rat prostate cancer cell proliferation appears to be mediated by functional androgen receptors (18,19). These characteristics support the use of prostate cancer cell lines as a model to assess the interactive roles of androgens and growth factors as modulators of prostate cancer cell function. As part of such analyses, we initiated studies to characterize AXC/SSh prostate cancer cell growth factor receptors and define growth factor modulation of cell function. Because 1) proliferation of some prostate cancer cells is androgen responsive (18), 2) normal rat ventral prostate contains TGF/3 receptors (6), and 3) TGF/3 is a hormonally regulated negative growth factor in human breast cancer cells (20), we chose to initially characterize AXC/SSh rat prostate cancer cell TGF/3 receptors. In this report we detail the results of our studies. Materials and Methods Cell lines and cell culture Clonally derived AXC/SSh rat prostate cancer cell lines C3, Dl, Tl, and T5 were used in these studies. Culture and subculture of these cells were previously described (16, 18). Materials FBS was obtained from HyClone Laboratories (Logan, UT), and other materials for cell culture were obtained and used as previously described (16, 18). Phenol red-free MEMEM without glutamine and sodium bicarbonate (Auto-Pow) was purchased from Flow Laboratories, Inc. (McLean, VA). Purified preparations of porcine TGF/31 and [125I]TGF/3 (-100 fiCi/^g) or bovine aFGF and bFGF were obtained from R & D Systems (Minneapolis, MN). Human recombinant IGF-I was obtained from Collaborative Research (Bedford, MA), and recombinant EGF and PDGF were purchased from Amgen Biologicals (Thousand Oaks, CA). Concentrated stock solutions of PDGF in 1.0 M acetic acid-0.15 M NaCl, or TGF/31 in 4 mM acetic acid-2 mg/ml BSA were stored at -20 C. Concentrated stock solutions of aFGF, bFGF, EGF, or IGF-I in 10 mM sodium phosphate, 0.1 M NaCl, and 0.05% ovalbumin, pH 7.5, were stored at -80 C. Mitogens were diluted into appropriate buffers immediately before use. [Metfry/-3H3]Thymidine was obtained from NEN/DuPont (Boston, MA). Disuccinimidyl suberate
819
was purchased from Pierce (Rockford, IL), and reagents for electrophoresis were from Bio-Rad (Richmond, CA). Aprotinin, pepstatin, phenylmethylsulfonylfluoride, ovalbumin, and radioinert thymidine were obtained from Sigma Chemical Co. (St. Louis, MO). Leupeptin was purchased from Calbiochem (La Jolla, CA). Other reagents were analytical grade. Solutions and culture media were prepared in water purified by reverse osmosis and then distilled from glass. TGFfi binding assays Prostate cancer cells, propagated on cell line-specific parental medium (16, 18), were subcultured on phenol red-free parental medium for 72 h. On the day before initiating the binding assay, these cells were harvested, and 105 cells/well were subcultured on 24-well plates in phenol red-free parental medium. At initiation of the binding assay, culture medium was removed and replaced with 1.0 ml binding buffer (40 mM HEPES and 1 mg/ml BSA in phenol red-free bicarbonate-containing MEMEM, pH 7.4) prewarmed to 37 C. After incubation at 37 C for 10 min, binding buffer was removed, and monolayers were washed a second time with 1 ml fresh binding buffer. At conclusion of the second 10-min 37 C incubation, the multiwell plates were cooled to 4 C. The binding buffer wash was then removed and replaced with 200 (A binding buffer containing 100-600 pM iodo-TGF/? prepared by adding the required quantity of radioinert TGF/? to stock 100 pM iodoTGF/3. Nonspecific binding was determined by incubating monolayers with 100 pM iodoTGF/3 and 20 nM radioinert TGF/3. Plates containing triplicate determinations were placed in sealed bags and incubated at 4 C for 4 h. At conclusion of this incubation, radiolabel containing binding buffer was removed, and monolayers were washed four times at 4 C with 1 ml cold binding buffer. Washed radiolabeled monolayers were solubilized by covering with 750 n\ 20 mM HEPES, pH 7.4, containing 1% Triton X-100 and 10% glycerol, followed by 30-min incubation at 37 C. After correction for nonspecific binding, data were analyzed by the method of Scatchard (21) and as double reciprocal plots (22). The mitogen specificity of iodo-TGF/3 binding was assessed by incubating parallel triplicate monolayer cultures with 100 pM iodo-TGF/3 in the absence or presence of 20 nM radioinert mitogen. To assess the effect of testosterone on T5 cell TGF/3 receptor content, cells cultured on parental medium were subcultured for 72 h on phenol red-free medium lacking testosterone. At 48 h before initiating quantification of TGF/3 receptors, 105 cells were plated/well on 24-well plates in phenol red-free medium which lacked testosterone or contained 10~7, 10~8, or 10~9 M testosterone. TGF/3 receptor site content was determined by saturation analysis, performed as detailed above. The isotope content of samples was quantified in a Beckman model 8500 7-scintillation spectrometer (Beckman, Palo Alto, CA). Polyacrylamide gel electrophoretic analysis of affinity-labeled TGFp receptors To affinity label TGF/3 receptors, parallel incubations were performed as detailed above using 100 pM iodo-TGF/3 in the absence or presence of 20 nM radioinert TGF/3. After 4-h incubation at 4 C, radioisotope-containing binding buffer was
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 19:32 For personal use only. No other uses without permission. . All rights reserved.
820
PROSTATE CANCER CELL TGF/3 RECEPTORS
removed, and monolayers were washed three times at 4 C using 2 ml fresh binding buffer for each wash. Washed monolayers were covered with fresh binding buffer, and sufficient freshly prepared disuccinimidyl suberate in dimethylsulfoxide was added to achieve 0.25 mM disuccinimidyl suberate and 1% dimethylsulfoxide. After 15-min incubation at 4 C, the crosslinking reaction was stopped by removing the incubation buffer and washing monolayers at 4 C with cell collection buffer. The collection buffer composition was 10 mM Tris-Cl, pH 7.0, containing 250 mM sucrose and 1 mM EDTA. Immediately before use, the collection buffer was supplemented to contain the following protease inhibitors: phenylmethylsulfonylfluoride, 0.5 mM; pepstatin, 1 /ng/ml; leupeptin, 1 Mg/ml; and aprotinin, 2.5 /zg/ml. Cells of washed monolayers were suspended by scraping into protease inhibitor containing collection buffer and collected by 5-min centrifugation at 12,000 X g at 4 C. Cell pellets were extracted by resuspension and incubation at 4 C in a minimum volume of protease inhibitor-supplemented cell collection buffer containing 1% Triton X-100. Iodo-TGFjS affinity-labeled proteins were characterized by polyacrylamide gel electrophoresis, performed essentially as detailed by Laemmli (23). Mitogen assays Prostate cancer cells were cultured on cell line-specific parental medium (16,18), and 5 X 104 cells/well were subcultured for 24 h on 24-well plates. These monolayers were washed with MEMEM, and growth was arrested by 24-h culture on serumfree MEMEM. To assess the effect of TGF/3 on prostate cancer cell DNA synthesis, medium was removed from growth-arrested cells and replaced with fresh MEMEM to which TGF0 in MEMEM was added to obtain the indicated final mitogen concentration. Culture was continued for 22 h, and DNA synthesis was assessed by adding serum-free MEMEM containing 1 MCi [3H]thymidine in 5 X 10~6 M thymidine. After 2-h culture at 37 C, the acid-insoluble radioisotope content of monolayers was determined as previously described (16). To determine the ability of TGF/3 to modulate bFGF mitogenicity in prostate cancer cells, growth-arrested monolayers were covered with fresh serum-free MEMEM containing the indicated concentration of TGF/3. After 3-h culture at 37 C, medium was supplemented to contain the indicated concentrations of bFGF, and culture was continued at 37 C for 22 h. DNA synthesis was then quantified as detailed above.
Endo • 1990 Voll26«No2
4-h incubation. Assessments of mitogen specificity of iodo-TGF/3 binding were performed by determining the ability of selected radioinert mitogens to inhibit iodoTGF0 binding to prostate cancer cells. These analyses showed (Table 1) that radioinert TGF/3 effectively inhibited iodo-TGF/3 binding to prostate cancer cells, whereas other prototypic mitogens were ineffective inhibitors. Titration analyses of iodo-TGF/3 binding by members of the C-, D-, and T-families of clonally derived prostate cancer cells showed that these cells contained limited capacity, high affinity binding sites (Fig. 1). Scatchard analysis of the binding data (Fig. 1) yielded a rectilinear TABLE 1. Mitogen specificity of TGF/3 binding to AXC/SSh rat prostate cancer cell TGF/3 receptors Competitor mitogen
Specific binding
remaining (%)
TGF/3 EGF FGF° IGF-I PDGF
0 103 104 109 116
Data are the mean for a single analysis, performed in triplicate, in which the radiolabeled TGF/3 concentration was 0.1 nM, and the radioinert mitogen concentration was 20 nM. a aFGF and bFGF gave identical results.
~ 0.75 -
0.50
0.25 -
Statistical analyses Significance of differences was assessed by multiway analysis of variance (24). 3.0
Results Characterization of TGF(3 binding to rat prostate cancer cells Preliminary analyses showed that monolayers of rat prostate cancer cells bound iodo-TGF/3 during incubation at 4 C and that maximum binding was achieved during
Bound ( fmol )
FIG. 1. Representative analyses of TGFjS binding to rat prostate cancer cell monolayers. T5 (•) or Tl (A) cells were cultured in 24-well plates and incubated with increasing concentrations of TGF/3 at 4 C for 4 h. Specific and nonspecific binding were quantified as detailed in the text. The inset shows saturation data used to construct the Scatchard representations.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 19:32 For personal use only. No other uses without permission. . All rights reserved.
PROSTATE CANCER CELL TGF/3 RECEPTORS plot consistent with the presence of a single class of binding sites. The results for T5 and T l cells (Fig. 1) are comparable to those obtained for identical analyses of C3 or Dl cells (data not shown). Comparative analysis of TGF/3 site content and dissociation constant showed that these essentially were identical for C3, Dl, and T5 cells (Table 2). In contrast, T l prostate cancer cell TGF/3 site content and dissociation constant were only 20-25% of those in the other cells (Table 2). Affinity labeling characterization of prostate cancer cell TGFp receptors Having established that rat prostate cancer cells contain limited capacity, high affinity TGF/3-binding sites (Table 2), we affinity labeled cell membrane binding components by using chemical cross-linking. These analyses showed that iodo-TGF/3 was bound to a component with an apparent mol wt of about 280 kDa (Fig. 2). The approximately 280-kDa binding species appeared to be the only TGF/3-binding component, because extended exposure of gels did not reveal other radiolabeled binding species (data not shown). Association of iodo-TGF/3 with the 280-kDa membrane component was highly specific, as shown by findings from parallel analyses that inclusion of a 200-fold molar excess of radioinert TGF/3 causes nearly quantitative elimination of iodo-TGF/3 binding (Fig. 2). Cross-linking of iodo-TGF/3 to identical numbers of T l or T5 cells showed that T l cells bound much less TGF/3 than did T5 cells (Fig. 2). This finding is consistent with the results of TGF/3 saturation analyses (Table 2 and Fig. 1), which demonstrated that T5 cell TGF0 receptor content is about 4-fold greater than that of T l cells. TGF/3 modulates Tl and T5 prostate cancer cell thymidine incorporation To assess functional status of prostate cancer cell TGF/3 receptors, we quantified TGF/3 modulation of T l and T5 prostate cancer cell thymidine incorporation. T5 and T l cell thymidine incorporation progressively decreased as culture medium TGF/3 content was incremenTABLE
2. TGF0 receptors of AXC/SSh rat prostate cancer cells Cell line C3 Dl Tl T5
Receptor content (sites/cell) 8,560 ± 1,450 13,160 ± 1,240 2,425 ± 490° 10,540 ± 1,025
Kd (PM)
160 ± 48 200 ± 53 24 ±3° 115 ± 15
Data are the mean ± SEM for two to six independent determinations, performed in triplicate. ° Significantly different from the value for all other cell lines, P < 0.05.
821
a
kDa
B
205-
103-
Cells:
T5
T1
FIG. 2. Representative polyacrylamide gel characterization of [125I] TGF/J affinity-labeled rat prostate cancer cells. Confluent monolayers were incubated for 4 h at 4 C with 0.1 nM [126I]TGF/3 in the absence (A and B) or presence (a and b) of 20 nM radioinert TGFjS. TGF/3 was cross-linked to receptors of washed monolayers by 15-min incubation at 4 C with buffer containing 0.25 mM disuccinimidyl suberate. Triton
X-100-soluble extracts were prepared and analyzed by electrophoresis on 5.5% polyacrylamide gels, which were dried and autoradiographed at —80 C. The position of migration of prestained myosin (205 kDa) and phosphorylase-b (103 kDa) are shown on the left. The top of the resolving gel is 10-11 mm above the band representing membrane protein-bound [125I]TGFi8.
tally increased from 1 to 100 pg/ml (Fig. 3). Both the magnitude and the rate of the TGF/3-mediated decrease in Tl and T5 cell thymidine incorporation were comparable (Fig. 3). A progressive increase in medium TGF/3 content from 0.1 to 10 ng/ml enhanced thymidine incorporation by Tl and T5 cells, and the rate of increase occurring with T l cells exceeded that with T5 cells (Fig. 3). The concentrations of TGF/3 required to achieve halfmaximum inhibition in Tl and T5 cells were not significantly different (Table 3). In contrast, the concentration of TGF/3 required to achieve half-maximum stimulation of T5 cell thymidine incorporation was approximately 9fold greater than that which caused half-maximum stimulation of T l cell thymidine incorporation (Table 3). TGF/3 modulates the ability of bFGF to affect T5 and Tl cell thymidine incorporation The capacity of TGF/3 to antagonize the mitogenic action of other growth factors is well established (25).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 19:32 For personal use only. No other uses without permission. . All rights reserved.
PROSTATE CANCER CELL TGF/3 RECEPTORS
822
Endo • 1990 Vol 126 • No 2
T1 Cells 25-
20-
=1 2 0.0025
0.01 0.05 0.25 1 TCF/3 Concentration (ng/ml)
JJQ
5
FIG. 3. TGF/3 is a bifunctional effector of Tl and T5 rat prostate cancer cell thymidine incorporation. (M) T5 or Tl (•) cells (5 x 104 cells/well) were cultured for 24 h on 24-well plates in serum-containing medium. Cells were growth arrested by subsequent 24-h culture on serum-free medium. Culture medium was replaced with serum-free medium containing the shown concentrations of TGF/3, and culture was continued for 22 h at which time [3H]thymidine incorporation was quantified. Significantly different from the value for T5 (*) or Tl (**) cells that did not receive TGF/3 (multiway analysis of variance, P < 0.05). Data are the mean ± SEM of two independent determinations, performed in duplicate. TABLE 3. Concentration of TGF/3 required for modulation of rat prostate cancer cell thymidine incorporation
Cell line Tl T5
TGF/3 required for half-maximum effect (ng/ml) Stimulation
Inhibition
0.36 ± 0.03 3.35 ± 0.15°
0.0027 ± 0.0009 0.0060 ± 0.0025
Data are the mean ± SEM for two independent determinations. Significantly different from the value for Tl cells, P < 0.05.
25-
T5 Cells
I 10
5-
0.02
0.1
0.2
1 2 bFCP (n^ml)
10
20
50
FIG. 4. TGFjS inhibits bFGF stimulation of Tl and T5 rat prostate cancer cell [3H]thymidine incorporation. Tl and T5 cells were cultured and growth arrested on 24-well plates as detailed in Fig. 3. Culture medium was then replaced with serum-free medium which either did not contain TGF/3 (S) or contained 0.01 (•) or 0.1 (•) ng TGF/3/ml. Cells were cultured for 3 h at 37 C, and wells then received the indicated amount of bFGF. After 22-h additional culture, thymidine incorporation was quantified. *, Significantly different from the value for cells that did not receive TGF/3 (multiway analysis of variance, P < 0.05). Data are the mean ± SEM for triplicate analyses.
0
Because we have shown that bFGF enhances rat prostate cancer cell thymidine incorporation (16), we sought to define the effect of TGF0 on bFGF mitogenicity in these cells. Titration analyses established that increasing medium bFGF content initially enhanced Tl or T5 cell thymidine incorporation and that maximum enhancement occurred when the medium bFGF content was 1-2 ng/ml (Fig. 4). Further increases in medium bFGF content caused progressive decrements in thymidine incorporation by T l or T5 cells (Fig. 4). Simultaneous inclusion of either 0.01 or 0.1 ng/ml TGF/3 significantly diminished the magnitude of bFGF stimulation of Tl or T5 cell thymidine incorporation. TGF/3-mediated inhibition of bFGF-enhanced prostate cancer cell thymidine incorporation was TGF/3 concentration dependent through that portion of the bFGF titration for which bFGF stimulated T l or T5 cell thymidine incorporation (Fig. 4). The concentration of bFGF required to achieve
TABLE 4. The effect of TGF/3 on the concentration of bFGF required for half-maximum stimulation of rat prostate cancer cell thymidine incorporation bFGF required for half-maximum stimulation (ng/ml) line
None
0.01 ng/ml TGF/3
0.10 ng/ml TGF/3
Tl T5
0.142 ± 0.019 0.083 ± 0.015
0.358 ± 0.093° 0.152 ± 0.002°
0.705 ±0.110 a ' 6 0.133 ± 0.011°
Data are the mean ± SEM for triplicate determinations. Significantly different from the value for similar cells which did not receive TGF0, P < 0.05. 6 Significantly different from the value for identically treated T5 cells, P < 0.05. 0
half-maximum stimulation of Tl cell thymidine incorporation significantly increased as the culture medium TGF/3 concentration was increased (Table 4). In contrast, although TGF/3 diminished the magnitude of the bFGF-mediated enhancement of T5 cell thymidine incorporation, the concentration of bFGF required to
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 19:32 For personal use only. No other uses without permission. . All rights reserved.
PROSTATE CANCER CELL TGF/3 RECEPTORS achieve half-maximum stimulation was independent of the medium TGF/3 content (Table 4). Androgens do not affect T5 prostate cancer cell TGF(3 receptor content We have shown that androgens inhibit in vitro proliferation of T5 prostate cancer cells and that the effects are mediated through functional androgen receptors (18, 19). To determine whether androgen effects may in part be mediated through changes in prostate cancer cell TGF/3 receptor content, we examined androgen modulation of T5 cell TGF/3 receptor content. These analyses demonstrated that physiological concentration of testosterone, which cause in vitro inhibition of T5 cell proliferation (18, 19), fail to affect T5 cell TGF/3 receptor content (Table 5). Our inability to detect an androgen effect on TGF/3 receptor content of these cells was not attributable to potential androgen-mediated changes in TGF/3 binding to its receptor, because we found that the Kd for TGF/3 binding to T5 cells is androgen concentration independent (Table 5). Discussion We show that AXC/SSh rat prostate cancer cells contain limited capacity, high affinity TGF/3-binding components which have mitogen binding specificity characteristic of TGF/3 receptors. T l cell TGF|3 receptor content is comparable to that reported (26) for cultured human prostate PC3 and DU145 cells which, respectively, contain 3,000 and 1,500 TGF0 receptors/cell. The TGF/3 receptor content of C3, Dl, or T5 cells is similar to that of osteoblasts (27, 28), NRK-49F (29, 30), or BALB/c 3T3 (29) cells which, respectively, contain 5,000-10,000, 17,000-19,000, or 16,000 TGF/3 receptors/ cell. In contrast, Swiss 3T3 cells contain 80,000 TGF,8 receptors/cell (31). The TGF/3 receptor content of normal rat ventral prostate membrane preparations is 50 fmol/mg DNA (6), approximately 300 sites/cell. Differences in the TGF/3 receptor content of normal rat ventral prostate (6) or rat prostate cancer cells may reflect differences in properties of the preparations analyzed. Osteoblast TGF/3 Kd has been reported to be 2.2 (26) or TABLE 5. AXC/SSh T5 rat prostate cancer cell TGF/3 receptor content is not modulated by testosterone Testosterone (M)
None 10"7 10"8 10"9
Sites/cell (% of control)
(pM)
Kd
100 118 ± 10 136 ± 19 108 ± 9
115 ± 15 96 ± 16 126 ± 1 110 ± 26
Data are the mean ± SEM for two independent determinations, performed in triplicate.
823
50-150 pM (27). In other cultured cells (25, 26, 29-31) or prostate membrane preparations (6) the Kd for TGF/3 ranges from 25-150 pM. These values are similar to the Kd of TGF/3 for rat prostate cancer cell receptors. The size of rat prostate cancer cell TGF/3 receptor (~280 kDa) is identical to that of normal rat ventral prostate (6) and the principal TGF/3 receptor of other mammalian cells (29, 31, 32). TGF/3 receptors ranging in size from 65-140 kDa have been described (27, 29, 31, 32). These species were not detected after chemical cross-linking of TGF/3 bound to rat prostate cancer cells (data not shown). TGF/3 is a bifunctional modulator of rat prostate cancer cell thymidine incorporation. Half-maximum inhibition of thymidine incorporation by T l or T5 cells requires 0.11-0.24 pM TGFjS, values at least 2 orders of magnitude less than that necessary to inhibit the function of numerous other cells. By example, half-maximum inhibition of soft agar colony formation (anchorage-independent growth) of multiple human carcinomas generally is achieved with 10-30 pM TGF/3 (33), whereas others suggest that half-maximum inhibition of colony formation of some human cells requires 320 pM TGF/3 (34). Similarly, enhancement of human osteosarcoma cell proliferation by either EGF or PDGF is inhibited by TGF/3 at a half-maximum concentration of 40 pM (35). TGF/3 inhibition of rat embryo fibroblast (36), NIH 3T3 (36), and NRK (37) colony formation, or primary rat hepatocyte thymidine incorporation (37) requires 1-15 pM TGF/3 for half-maximum effect. In contrast, halfmaximum inhibition of EGF-stimulated mink lung cell proliferation requires 300 pM TGF/3 (38). The concentration of TGF/3 required for half-maximum stimulation of Tl cell thymidine incorporation was approximately 9-fold greater than that needed to achieve the same effect in T5 cells. Because identical numbers of T l and T5 cells were growth arrested, and plating efficiency and survival of these cells are comparable, the difference in T l and T5 cell sensitivity to TGF/3-modulated thymidine incorporation would not appear to be attributable to cell density-mediated differences in TGF/3 sensitivity, as is characteristic of other cells (28, 39-41). Our studies do not unequivocally eliminate the possibility that the apparent greater sensitivity of Tl cells to TGF/3enhanced thymidine incorporation is due to elevated TGF(8 secretion. If this occurred, the quantity of TGF/3 required to achieve a comparable increase in T5 cell thymidine incorporation would be artificially elevated compared to that in Tl cells. This seems an unlikely explanation. If Tl cells produced more TGF/3 than T5 cells, the increased TGF/3 content of Tl cell cultures would cause Tl cells to appear more sensitive to TGF/3 inhibition of thymidine incorporation. This is not the case. Finally, the concentrations of TGF/3 required for
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 19:32 For personal use only. No other uses without permission. . All rights reserved.
824
PROSTATE CANCER CELL TGF/5 RECEPTORS
half-maximum stimulation of Tl and T5 cell thymidine incorporation are representative of the range of concentrations required for half-maximum increase in colony formation (33, 34,42), proliferation (28), or EGF receptor content (41) of other cells. Concentration-dependent bifunctional TGF/3 modulation of T l and T5 cell thymidine incorporation appears to be unique to these cells. Although a considerable literature establishes that TGF0 may either stimulate or inhibit cell function (27, 28, 33-42), both responses rarely affect the same function in a cell. In the known cases the response is modulated by other effectors. For example, TGF/3 inhibits EGF-mediated and potentiates PDGFmediated colony formation by Myc-1 cells (33). Similarly, the ability of TGF/3 to either decrease (40) or increase (41) NRK cell EGF receptor content is dependent on cell density; the former occurs only in sparse cultures, whereas the latter occurs only in confluent cultures. Moreover, TGF/3 enhancement of NRK cell EGF receptor content occurs as a consequence of elevated receptor synthesis (41), whereas the decrease in apparent EGF receptor content is probably due to changes in receptor Kd for EGF (40). The dose-response curve for bFGF modulation of Tl or T5 cell thymidine incorporation was bell shaped, demonstrating that high concentrations of bFGF depress thymidine incorporation by these cells. This finding essentially is identical to results we detailed in prior studies (16). TGF/3 inhibited bFGF enhancement of Tl and T5 cell thymidine incorporation, and the effect was TGF/3 concentration dependent through that range of the bFGF titration which produced enhancement of thymidine incorporation. The concentration of bFGF required to achieve half-maximum increase in T5 cell thymidine incorporation was independent of culture medium TGF/3 content and essentially identical to the value we previously reported (16). In contrast, the concentration of bFGF required to achieve a half-maximum increase in T l cell thymidine incorporation increased 5-fold as the culture medium TGF/3 concentration increased. These differences in the effect of TGF/3 on bFGF modulation of Tl and T5 cell thymidine incorporation imply potential differences in the mode of action by TGF/3 as a modulator of bFGF action in these cells. The fact that the concentration of bFGF necessary for half-maximum increase in T5 cell thymidine incorporation is independent of medium TGF(3 content suggests that TGF/3 affects T5 cell function at a point distal to bFGF interaction with T5 cell FGF receptors. This relationship is characteristic of TGF/? modulation of EGF-stimulated mink lung cell proliferation (38) or NRK cell EGF receptor content (41). In contrast, the fact that the concentration of bFGF required to achieve a half-maximum increase in Tl cell thymidine incorporation increases as the medium
Endo • 1990 Voll26«No2
TGF/3 content is increased implies that TGF/3 proximally affects bFGF function in T l cells. The half-maximum concentration of TGF/3 necessary to inhibit bFGF stimulation of bovine aortic arch (43, 44) or retinal capillary (45) endothelial cell proliferation is 1-30 pM. Thus, sensitivity of these cells to TGFjS inhibition of function is significantly less than that of rat prostate cancer cells. Because the bFGF concentration required for a half-maximum response was independent of medium TGF/3 content (43, 44), the effects of TGF/3 in these cells are distal to the bFGF receptor. Consequently, the mechanism of TGF/3 modulation of endothelial cell and T5 prostate cancer cell bFGF response may be similar. Results of our studies establish that rat prostate cancer cells contain functional TGF/3 receptors and imply the presence of functional bFGF receptors. The finding that TGFjS modulates bFGF effects establishes that mitogen modulation of prostate cancer cell function is multifactorial. The bifunctional TGF/3 modulation of prostate cancer cell DNA synthesis is unique and provides some insight into the potential complexity of mitogen modulation of prostate cancer cell proliferation. The mechanism by which these mitogens interact to affect cancer cell function is unknown. However, the differences we detail for T l and T5 cells implies that some interactive effects of mitogens may be cell line specific.
References 1. Griffin JE, Wilson JD 1980 The syndromes of androgen resistance. N Engl J Med 302:198 2. Huggins C, Hodges CV 1941 Studies in prostatic cancer. I. The effects of castration, of estrogen and of androgen injection on serum phosphatases in metastatic carcinoma of the prostate. Cancer Res 1:293 3. McKeehan WL, Adams PS, Rosser MP 1984 Direct mitogenic effects of insulin, epidermal growth factor, 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 4. Eaton CL, Davies P, Phillips MEA1988 Growth factor involvement and oncogene expression in prostatic tumours. J Steroid Biochem 30:341 5. Traish AM, Wotiz HH 1987 Prostatic epidermal growth factor receptors and their regulation by androgens. Endocrinology 121:1461 6. Kyprianou N, Isaacs JT 1988 Identification of a cellular receptor for transforming growth factor-/? in rat ventral prostate and its negative regulation by androgens. Endocrinology 123:2124 7. Davies P, Eaton CL 1989 Binding of epidermal growth factor by human normal, hypertrophic, and carcinomatous prostate. Prostate 14:123 8. Schuurmans ALG, Bolt J, Mulder E 1988 Androgens stimulate both growth rate and epidermal growth factor receptor activity of the human prostate tumor cell LNCaP. Prostate 12:55 9. Jacobs SC, Story MT, Sasse J, Lawson RK 1988 Characterization of growth factors derived from the rat ventral prostate. J Urol 139:1106 10. Maehama S, Li D, Nanri H, Leykam JF, Deuel TF 1986 Purification and partial characterization of prostate-derived growth factor. Proc Natl Acad Sci USA 83:8162 11. Ikeda T, Lioubin MN, Marquardt H 1987 Human transforming
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 19:32 For personal use only. No other uses without permission. . All rights reserved.
PROSTATE CANCER CELL TGF/3 RECEPTORS
12.
13. 14. 15.
16. 17. 18. 19. 20.
21. 22. 23. 24. 25. 26. 27.
28.
growth factor type /32: production by a prostatic adenocarcinoma cell line, purification, and initial characterization. Biochemistry 26:2406 Story MT, Esch F, Shimasaki S, Sasse J, Jacobs SC, Lawson RK 1987 Amino-terminal sequence of a large form of basic fibroblast growth factor isolated from human benign prostatic hyperplastic tissue. Biochem Biophys Res Commun 142:702 Story MT, Sasse J, Jacobs SC, Lawson RK 1987 Prostatic growth factor: purification and structural relationship to basic fibroblast growth factor. Biochemistry 26:3843 Mydlo JH, Michaeli J, Heston WDW, Fair WR 1988 Expression of basic fibroblast growth factor mRNA in benign prostatic hyperplasia and prostatic carcinoma. Prostate 13:241 Matuo Y, Nishi N, Matsui S-i, Sandberg AA, Isaacs JT, Wada F 1987 Heparin binding affinity of rat prostatic growth factor in normal and cancerous prostates: partial purification and characterization of rat prostatic growth factor in the Dunning tumor. Cancer Res 47:188 Shain SA, Koger JD 1989 Differences in responsiveness of clonally derived AXC/SSh rat prostate cancer cells to secreted or prototypic mitogens. Cancer Res 49:3898 Shain SA, Huot RI, Gorelic LS, Smith GC 1984 Biochemical and morphological characterization of clonal AXC rat prostate cancer cells. Cancer Res 44:2033 Huot RI, Shain SA 1986 Differential androgen modulation of AXC/ SSh rat prostate cancer cell proliferation in vitro and its antagonism by antiandrogen. Cancer Res 46:3775 Lin AL, Shain SA 1989 Androgen modulation of prostate cancer cell androgen receptor content is cell line specific. Mol Cell Endocrinol 63:75 Knabbe C, Lippman ME, Wakefield LM, Flanders KC, Kasid A, Derynck R, Dickson RB 1987 Evidence that transforming growth factor-0 is a hormonally regulated negative growth factor in human breast cancer cells. Cell 48:417 Scatchard G 1949 The attractions of proteins for small molecules and ions. Ann NY Acad Sci 51:660 Shain SA, Boesel RW 1978 Human prostate steroid hormone receptor quantitation. Current methodology and possible utility as a clinical discriminant in carcinoma. Invest Urol 16:169 Laemmli UK 1970 Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (Lond) 227:680 Nie NH, Hull CH, Jenkins JG, Steinbrenner K, Bent DH 1975 SPSS: Statistical Package for the Social Sciences, ed 2. McGrawHill, New York Sporn MB, Roberts AB, Wakefield LM, de Crombrugghe B 1987 Some recent advances in the chemistry and biology of transforming growth factor-beta. J Cell Biol 105:1039 Wilding G, Zugmeier G, Knabbe C, Flanders K, Gelmann E 1989 Differential effects of transforming growth factor /3 on human prostate cancer cells in vitro. Mol Cell Endocrinol 62:79 Centrella M, McCarthy TL, Canalis E 1988 Parathyroid hormone modulates transforming growth factor /? activity and binding in osteoblast-enriched cell cultures from fetal rat parietal bone. Proc Natl Acad Sci USA 85:5889 Robey PG, Young MF, Flanders KC, Roche NS, Kondaiah P, Reddi AH, Termine JD, Sporn MB, Roberts AB 1987 Osteoblasts synthesize and respond to transforming growth factor-type 0 (TGF-/3) in vitro. J Cell Biol 105:457
825
29. Massague J, Like B 1985 Cellular receptors for type /? transforming growth factor. Ligand binding and affinity labeling in human and rodent cell lines. J Biol Chem 260:2636 30. Frolik CA, Wakefield LM, Smith DM, Sporn MB 1984 Characterization of a membrane receptor for transforming growth factor-/? in normal rat kidney fibroblasts. J Biol Chem 259:10995 31. Fanger BO, Wakefield LM, Sporn MB 1986 Structure and properties of the cellular receptor for transforming growth factor type /3. Biochemistry 25:3083 32. Chiefetz S, Like B, Massague J 1986 Cellular distribution of type I and type II receptors for transforming growth factor-/9. J Biol Chem 261:9972 33. Roberts AB, Anzano MA, Wakefield LM, Roche NS, Stern DF, Sporn MB 1985 Type /? transforming growth factor: a bifunctional regulator of cellular growth. Proc Natl Acad Sci USA 82:119 34. Moses HL, Tucker RF, Leof EB, Coffey Jr RJ, Halper J, Shipley GD 1985 Type-/? transforming growth factor is a growth stimulator and a growth inhibitor. In: Feramisco J, Ozanne B, Stiles C (eds) Growth Factors and Transformation. Cold Spring Harbor Laboratory, Cold Spring Harbor, p. 65 35. Mioh H, Chen J-K 1989 Differential inhibitory effects of TGF-/3 on EGF, PDGF-, and HBFG-1-stimulated MG63 human osteosarcoma cell growth: possible involvement of growth factor interactions at the receptor and postreceptor levels. J Cell Physiol 139:509 36. Anzano MA, Roberts AB, Sporn MB 1986 Anchorage-independent growth of primary rat embryo cells is induced by platelet-derived growth factor and inhibited by type-beta transforming growth factor. J Cell Physiol 126:312 37. Nakamura T, Tomita Y, Hirai R, Yamaoka K, Kaji K, Ichihara A 1985 Inhibitory effect of transforming growth factor-/? on DNA synthesis of adult rat hepatocytes in primary culture. Biochem Biophys Research Commun 133:1042 38. Like B, Massague J 1986 The antiproliferative effect of type /3 transforming growth factor occurs at a level distal from receptors for growth-activating factors. J Biol Chem 261:13426 39. Centrella M, McCarthy TL, Canalis E 1987 Transforming growth factor /3 is a bifunctional regulator of replication and collagen synthesis in osteoblast-enriched cell cultures from fetal rat bone. J Biol Chem 262:2869 40. Massague J 1985 Transforming growth factor-/3 modulates the high-affinity receptors for epidermal growth factor and transforming growth factor-a. J Cell Biol 100:1508 41. Assoian RK, Frolik CA, Roberts AB, Miller DM, Sporn MB 1984 Transforming growth factor-/? controls receptor levels for epidermal growth factor in NRK fibroblasts. Cell 36:35 42. Tucker RF, Shipley GD, Moses HL, Holley RW 1984 Growth inhibitor from BSC-1 cells closely related to platelet type /? transforming growth factor. Science 226:705 43. Frater-Schroder M, Muller G, Birchmeier W, Bohlen P 1986 Transforming growth factor-beta inhibits endothelial cell proliferation. Biochem Biophys Res Commun 137:295 44. Baird A, Durkin T 1986 Inhibition of endothelial cell proliferation by type ^-transforming growth factor: interactions with acidic and basic fibroblast growth factors. Biochem Biophys Res Commun 138:476 45. Bensaid M, Malecaze F, Bayard F, Tauber JP 1989 Opposing effects of basic fibroblast growth factor and transforming growth factor-/? on the proliferation of cultured bovine retinal capillary endothelial (BREC) cells. Exp Eye Res 48:791
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 19:32 For personal use only. No other uses without permission. . All rights reserved.
Vol. 126, No. 2 Printed in U.S.A.
Rat Prostate Cancer Cells Contain Functional Receptors for Transforming Growth Factor-/?* SYDNEY A. SHAIN, ALAN L. LIN, JUDITH D. ROGER, AND ANDREW G. KARAGANIS Department of Cellular and Molecular Biology, Southwest Foundation for Biomedical Research, San Antonio, Texas 78284
ABSTRACT. Clonally derived C-, D-, and T-family AXC/SSh
of function is comparable to that of other mammalian cells.
rat prostate cancer cell lines contain transforming growth factor|3 (TGF/3) receptors. The content in C3, Dl, Tl, and T5 cells, respectively, was 8,560 ± 1,450,13,160 ± 1,240, 2,425 ± 490, and 10,540 ± 1,025 sites/cell (mean ± SEM). Respective Kd values were 160 ± 48, 200 ± 53, 24 ± 3, and 115 ± 15 pM (mean ± SEM). T l cell TGF/3 receptor site content and Kd differed significantly from those of other prostate cancer cell lines (P < 0.05). TGF/8 is a bifunctional concentration-dependent modulator of T l and T5 cell thymidine incorporation. At low concentrations, thymidine incorporation was inhibited, whereas as the medium TGFjS content was increased, Tl and T5 cell thymidine incorporation was stimulated. The concentrations of TGF/3 causing half-maximum inhibition of T l or T5 cell thymidine incorporation, respectively, were 0.11 and 0.24 pM, whereas the respective TGF/3 concentrations causing half-maximum stimulation of thymidine incorporation were 14.4 and 134 pM. These findings establish that rat prostate cancer cell sensitivity to TGF/3 inhibition of function is at least 2 orders of magnitude greater than that of most other mammalian cells. In contrast, the sensitivity of rat prostate cancer cells to TGF/3 enhancement
TGF/3 inhibited basic fibroblast growth factor (bFGF) stimulation of T l and T5 cell thymidine incorporation. Because the concentration of bFGF required for half-maximum increase of T5 cell thymidine incorporation was independent of medium TGF/3 content, the effect of TGF/3 is distal to the T5 cell bFGF receptor. In contrast, the concentration of bFGF required for half-maximum increase in T l cell thymidine incorporation increased 5-fold as the medium TGF/3 content was increased; suggesting that the effect of TGF/3 in Tl cells is proximal to the Tl cell bFGF receptor. Our studies establish that rat prostate cancer cells contain functional TGF/3 receptors, imply the presence of functional bFGF receptors, and demonstrate that mitogen modulation of prostate cancer cell function is multifactorial. The finding that TGF/3 is a bifunctional effector of prostate cancer cell DNA synthesis provides some insight into the potential complexity of mitogen modulation of prostate cancer cell proliferation. The mechanism by which these mitogens interact is unknown; however, our studies suggest that some interactive effects may be cell line specific. {Endocrinology 126: 818-825, 1990)
T
HE roles of androgens as principal effectors of prostate differentiation (1) and modulators of normal (1) or neoplastic cell function (2) are well established. However, results of recent studies suggest that growth factors also may importantly affect normal and neoplastic prostate cell function. By example, in vitro studies show that EGF1 and insulin enhance the proliferation of normal rat ventral prostate epithelium (3) and
Received August 21, 1989. Address all correspondence and requests for reprints to: Dr.. Sydney A. Shain, Department of Obstetrics and Gynecology, University of Texas Health Science Center, 7703 Floyd Curl Drive, San Antonio, Texas 78284. * This work was supported in part by Grant DK-39766 from the DHHS. 1 The following abbreviations are used: EGF, epidermal growth factor; TGFa, transforming growth factor-a; TGF/3, transforming growth factor-/3; TGF/31, homodimeric transforming growth factor-/31; aFGF, acidic fibroblast growth factor; bFGF, basic fibroblast growth factor; ECGF, endothelial cell growth factor; 5aDHT, 5a-dihydrotestosterone; FBS, fetal bovine serum; IGF-I, insulin-like growth factorI; PDGF, platelet-derived growth factor; MEMEM, Modified Eagle's Minimum Essential Medium.
that EGF and TGFa enhance the proliferation of normal canine prostate epithelium (4). Rat ventral prostate contains EGF (5) and TGF/3 (6) receptors, and cell receptor content is androgen modulated. Human normal (7), hypertrophic (7), and carcinomatous prostate cells (7, 8) also contain EGF receptors. An EGF-like mitogen (9), bFGF (9), and a 25-kDa acid-stable mitogen lacking transforming activity (10) are produced by normal rat ventral prostate. Human prostate carcinoma PC-3 cells elaborate TGF/3 (11), an amino-terminal extended bFGF has been purified from human benign hyperplastic tissue (12, 13), and bFGF transcripts are detectable in specimens of human prostatic carcinoma (14). Dunning rat prostate carcinoma elaborates a heparin-binding mitogen with properties characteristic of ECGF (15), and we have shown that AXC/SSh rat prostate carcinoma cells secrete mitogens with properties characteristic of the FGF family of growth factors (16). These findings suggest potential interactive roles for androgens and growth factors as modulators of normal and neoplastic prostate cell
818 The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 19:32 For personal use only. No other uses without permission. . All rights reserved.
PROSTATE CANCER CELL TGF0 RECEPTORS function. We have described clonally derived AXC/SSh rat prostate cancer cells which are distinguished as C-, D-, or Tfamily members based on their maintenance on medium containing charcoal-stripped serum which, respectively, has no added steroid or contains either 10~7 M 5«DHT or 10~7 M testosterone (17, 18). We have shown that in vitro proliferation of C- and D-family member cell lines is not androgen modulated (18), that androgens enhance T l cell and diminish T5 cell proliferation in vitro (18), and that the response of AXC/SSh prostate cancer cells to secreted or prototypic mitogens differs (16). Androgen modulation of rat prostate cancer cell proliferation appears to be mediated by functional androgen receptors (18,19). These characteristics support the use of prostate cancer cell lines as a model to assess the interactive roles of androgens and growth factors as modulators of prostate cancer cell function. As part of such analyses, we initiated studies to characterize AXC/SSh prostate cancer cell growth factor receptors and define growth factor modulation of cell function. Because 1) proliferation of some prostate cancer cells is androgen responsive (18), 2) normal rat ventral prostate contains TGF/3 receptors (6), and 3) TGF/3 is a hormonally regulated negative growth factor in human breast cancer cells (20), we chose to initially characterize AXC/SSh rat prostate cancer cell TGF/3 receptors. In this report we detail the results of our studies. Materials and Methods Cell lines and cell culture Clonally derived AXC/SSh rat prostate cancer cell lines C3, Dl, Tl, and T5 were used in these studies. Culture and subculture of these cells were previously described (16, 18). Materials FBS was obtained from HyClone Laboratories (Logan, UT), and other materials for cell culture were obtained and used as previously described (16, 18). Phenol red-free MEMEM without glutamine and sodium bicarbonate (Auto-Pow) was purchased from Flow Laboratories, Inc. (McLean, VA). Purified preparations of porcine TGF/31 and [125I]TGF/3 (-100 fiCi/^g) or bovine aFGF and bFGF were obtained from R & D Systems (Minneapolis, MN). Human recombinant IGF-I was obtained from Collaborative Research (Bedford, MA), and recombinant EGF and PDGF were purchased from Amgen Biologicals (Thousand Oaks, CA). Concentrated stock solutions of PDGF in 1.0 M acetic acid-0.15 M NaCl, or TGF/31 in 4 mM acetic acid-2 mg/ml BSA were stored at -20 C. Concentrated stock solutions of aFGF, bFGF, EGF, or IGF-I in 10 mM sodium phosphate, 0.1 M NaCl, and 0.05% ovalbumin, pH 7.5, were stored at -80 C. Mitogens were diluted into appropriate buffers immediately before use. [Metfry/-3H3]Thymidine was obtained from NEN/DuPont (Boston, MA). Disuccinimidyl suberate
819
was purchased from Pierce (Rockford, IL), and reagents for electrophoresis were from Bio-Rad (Richmond, CA). Aprotinin, pepstatin, phenylmethylsulfonylfluoride, ovalbumin, and radioinert thymidine were obtained from Sigma Chemical Co. (St. Louis, MO). Leupeptin was purchased from Calbiochem (La Jolla, CA). Other reagents were analytical grade. Solutions and culture media were prepared in water purified by reverse osmosis and then distilled from glass. TGFfi binding assays Prostate cancer cells, propagated on cell line-specific parental medium (16, 18), were subcultured on phenol red-free parental medium for 72 h. On the day before initiating the binding assay, these cells were harvested, and 105 cells/well were subcultured on 24-well plates in phenol red-free parental medium. At initiation of the binding assay, culture medium was removed and replaced with 1.0 ml binding buffer (40 mM HEPES and 1 mg/ml BSA in phenol red-free bicarbonate-containing MEMEM, pH 7.4) prewarmed to 37 C. After incubation at 37 C for 10 min, binding buffer was removed, and monolayers were washed a second time with 1 ml fresh binding buffer. At conclusion of the second 10-min 37 C incubation, the multiwell plates were cooled to 4 C. The binding buffer wash was then removed and replaced with 200 (A binding buffer containing 100-600 pM iodo-TGF/? prepared by adding the required quantity of radioinert TGF/? to stock 100 pM iodoTGF/3. Nonspecific binding was determined by incubating monolayers with 100 pM iodoTGF/3 and 20 nM radioinert TGF/3. Plates containing triplicate determinations were placed in sealed bags and incubated at 4 C for 4 h. At conclusion of this incubation, radiolabel containing binding buffer was removed, and monolayers were washed four times at 4 C with 1 ml cold binding buffer. Washed radiolabeled monolayers were solubilized by covering with 750 n\ 20 mM HEPES, pH 7.4, containing 1% Triton X-100 and 10% glycerol, followed by 30-min incubation at 37 C. After correction for nonspecific binding, data were analyzed by the method of Scatchard (21) and as double reciprocal plots (22). The mitogen specificity of iodo-TGF/3 binding was assessed by incubating parallel triplicate monolayer cultures with 100 pM iodo-TGF/3 in the absence or presence of 20 nM radioinert mitogen. To assess the effect of testosterone on T5 cell TGF/3 receptor content, cells cultured on parental medium were subcultured for 72 h on phenol red-free medium lacking testosterone. At 48 h before initiating quantification of TGF/3 receptors, 105 cells were plated/well on 24-well plates in phenol red-free medium which lacked testosterone or contained 10~7, 10~8, or 10~9 M testosterone. TGF/3 receptor site content was determined by saturation analysis, performed as detailed above. The isotope content of samples was quantified in a Beckman model 8500 7-scintillation spectrometer (Beckman, Palo Alto, CA). Polyacrylamide gel electrophoretic analysis of affinity-labeled TGFp receptors To affinity label TGF/3 receptors, parallel incubations were performed as detailed above using 100 pM iodo-TGF/3 in the absence or presence of 20 nM radioinert TGF/3. After 4-h incubation at 4 C, radioisotope-containing binding buffer was
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 19:32 For personal use only. No other uses without permission. . All rights reserved.
820
PROSTATE CANCER CELL TGF/3 RECEPTORS
removed, and monolayers were washed three times at 4 C using 2 ml fresh binding buffer for each wash. Washed monolayers were covered with fresh binding buffer, and sufficient freshly prepared disuccinimidyl suberate in dimethylsulfoxide was added to achieve 0.25 mM disuccinimidyl suberate and 1% dimethylsulfoxide. After 15-min incubation at 4 C, the crosslinking reaction was stopped by removing the incubation buffer and washing monolayers at 4 C with cell collection buffer. The collection buffer composition was 10 mM Tris-Cl, pH 7.0, containing 250 mM sucrose and 1 mM EDTA. Immediately before use, the collection buffer was supplemented to contain the following protease inhibitors: phenylmethylsulfonylfluoride, 0.5 mM; pepstatin, 1 /ng/ml; leupeptin, 1 Mg/ml; and aprotinin, 2.5 /zg/ml. Cells of washed monolayers were suspended by scraping into protease inhibitor containing collection buffer and collected by 5-min centrifugation at 12,000 X g at 4 C. Cell pellets were extracted by resuspension and incubation at 4 C in a minimum volume of protease inhibitor-supplemented cell collection buffer containing 1% Triton X-100. Iodo-TGFjS affinity-labeled proteins were characterized by polyacrylamide gel electrophoresis, performed essentially as detailed by Laemmli (23). Mitogen assays Prostate cancer cells were cultured on cell line-specific parental medium (16,18), and 5 X 104 cells/well were subcultured for 24 h on 24-well plates. These monolayers were washed with MEMEM, and growth was arrested by 24-h culture on serumfree MEMEM. To assess the effect of TGF/3 on prostate cancer cell DNA synthesis, medium was removed from growth-arrested cells and replaced with fresh MEMEM to which TGF0 in MEMEM was added to obtain the indicated final mitogen concentration. Culture was continued for 22 h, and DNA synthesis was assessed by adding serum-free MEMEM containing 1 MCi [3H]thymidine in 5 X 10~6 M thymidine. After 2-h culture at 37 C, the acid-insoluble radioisotope content of monolayers was determined as previously described (16). To determine the ability of TGF/3 to modulate bFGF mitogenicity in prostate cancer cells, growth-arrested monolayers were covered with fresh serum-free MEMEM containing the indicated concentration of TGF/3. After 3-h culture at 37 C, medium was supplemented to contain the indicated concentrations of bFGF, and culture was continued at 37 C for 22 h. DNA synthesis was then quantified as detailed above.
Endo • 1990 Voll26«No2
4-h incubation. Assessments of mitogen specificity of iodo-TGF/3 binding were performed by determining the ability of selected radioinert mitogens to inhibit iodoTGF0 binding to prostate cancer cells. These analyses showed (Table 1) that radioinert TGF/3 effectively inhibited iodo-TGF/3 binding to prostate cancer cells, whereas other prototypic mitogens were ineffective inhibitors. Titration analyses of iodo-TGF/3 binding by members of the C-, D-, and T-families of clonally derived prostate cancer cells showed that these cells contained limited capacity, high affinity binding sites (Fig. 1). Scatchard analysis of the binding data (Fig. 1) yielded a rectilinear TABLE 1. Mitogen specificity of TGF/3 binding to AXC/SSh rat prostate cancer cell TGF/3 receptors Competitor mitogen
Specific binding
remaining (%)
TGF/3 EGF FGF° IGF-I PDGF
0 103 104 109 116
Data are the mean for a single analysis, performed in triplicate, in which the radiolabeled TGF/3 concentration was 0.1 nM, and the radioinert mitogen concentration was 20 nM. a aFGF and bFGF gave identical results.
~ 0.75 -
0.50
0.25 -
Statistical analyses Significance of differences was assessed by multiway analysis of variance (24). 3.0
Results Characterization of TGF(3 binding to rat prostate cancer cells Preliminary analyses showed that monolayers of rat prostate cancer cells bound iodo-TGF/3 during incubation at 4 C and that maximum binding was achieved during
Bound ( fmol )
FIG. 1. Representative analyses of TGFjS binding to rat prostate cancer cell monolayers. T5 (•) or Tl (A) cells were cultured in 24-well plates and incubated with increasing concentrations of TGF/3 at 4 C for 4 h. Specific and nonspecific binding were quantified as detailed in the text. The inset shows saturation data used to construct the Scatchard representations.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 19:32 For personal use only. No other uses without permission. . All rights reserved.
PROSTATE CANCER CELL TGF/3 RECEPTORS plot consistent with the presence of a single class of binding sites. The results for T5 and T l cells (Fig. 1) are comparable to those obtained for identical analyses of C3 or Dl cells (data not shown). Comparative analysis of TGF/3 site content and dissociation constant showed that these essentially were identical for C3, Dl, and T5 cells (Table 2). In contrast, T l prostate cancer cell TGF/3 site content and dissociation constant were only 20-25% of those in the other cells (Table 2). Affinity labeling characterization of prostate cancer cell TGFp receptors Having established that rat prostate cancer cells contain limited capacity, high affinity TGF/3-binding sites (Table 2), we affinity labeled cell membrane binding components by using chemical cross-linking. These analyses showed that iodo-TGF/3 was bound to a component with an apparent mol wt of about 280 kDa (Fig. 2). The approximately 280-kDa binding species appeared to be the only TGF/3-binding component, because extended exposure of gels did not reveal other radiolabeled binding species (data not shown). Association of iodo-TGF/3 with the 280-kDa membrane component was highly specific, as shown by findings from parallel analyses that inclusion of a 200-fold molar excess of radioinert TGF/3 causes nearly quantitative elimination of iodo-TGF/3 binding (Fig. 2). Cross-linking of iodo-TGF/3 to identical numbers of T l or T5 cells showed that T l cells bound much less TGF/3 than did T5 cells (Fig. 2). This finding is consistent with the results of TGF/3 saturation analyses (Table 2 and Fig. 1), which demonstrated that T5 cell TGF0 receptor content is about 4-fold greater than that of T l cells. TGF/3 modulates Tl and T5 prostate cancer cell thymidine incorporation To assess functional status of prostate cancer cell TGF/3 receptors, we quantified TGF/3 modulation of T l and T5 prostate cancer cell thymidine incorporation. T5 and T l cell thymidine incorporation progressively decreased as culture medium TGF/3 content was incremenTABLE
2. TGF0 receptors of AXC/SSh rat prostate cancer cells Cell line C3 Dl Tl T5
Receptor content (sites/cell) 8,560 ± 1,450 13,160 ± 1,240 2,425 ± 490° 10,540 ± 1,025
Kd (PM)
160 ± 48 200 ± 53 24 ±3° 115 ± 15
Data are the mean ± SEM for two to six independent determinations, performed in triplicate. ° Significantly different from the value for all other cell lines, P < 0.05.
821
a
kDa
B
205-
103-
Cells:
T5
T1
FIG. 2. Representative polyacrylamide gel characterization of [125I] TGF/J affinity-labeled rat prostate cancer cells. Confluent monolayers were incubated for 4 h at 4 C with 0.1 nM [126I]TGF/3 in the absence (A and B) or presence (a and b) of 20 nM radioinert TGFjS. TGF/3 was cross-linked to receptors of washed monolayers by 15-min incubation at 4 C with buffer containing 0.25 mM disuccinimidyl suberate. Triton
X-100-soluble extracts were prepared and analyzed by electrophoresis on 5.5% polyacrylamide gels, which were dried and autoradiographed at —80 C. The position of migration of prestained myosin (205 kDa) and phosphorylase-b (103 kDa) are shown on the left. The top of the resolving gel is 10-11 mm above the band representing membrane protein-bound [125I]TGFi8.
tally increased from 1 to 100 pg/ml (Fig. 3). Both the magnitude and the rate of the TGF/3-mediated decrease in Tl and T5 cell thymidine incorporation were comparable (Fig. 3). A progressive increase in medium TGF/3 content from 0.1 to 10 ng/ml enhanced thymidine incorporation by Tl and T5 cells, and the rate of increase occurring with T l cells exceeded that with T5 cells (Fig. 3). The concentrations of TGF/3 required to achieve halfmaximum inhibition in Tl and T5 cells were not significantly different (Table 3). In contrast, the concentration of TGF/3 required to achieve half-maximum stimulation of T5 cell thymidine incorporation was approximately 9fold greater than that which caused half-maximum stimulation of T l cell thymidine incorporation (Table 3). TGF/3 modulates the ability of bFGF to affect T5 and Tl cell thymidine incorporation The capacity of TGF/3 to antagonize the mitogenic action of other growth factors is well established (25).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 19:32 For personal use only. No other uses without permission. . All rights reserved.
PROSTATE CANCER CELL TGF/3 RECEPTORS
822
Endo • 1990 Vol 126 • No 2
T1 Cells 25-
20-
=1 2 0.0025
0.01 0.05 0.25 1 TCF/3 Concentration (ng/ml)
JJQ
5
FIG. 3. TGF/3 is a bifunctional effector of Tl and T5 rat prostate cancer cell thymidine incorporation. (M) T5 or Tl (•) cells (5 x 104 cells/well) were cultured for 24 h on 24-well plates in serum-containing medium. Cells were growth arrested by subsequent 24-h culture on serum-free medium. Culture medium was replaced with serum-free medium containing the shown concentrations of TGF/3, and culture was continued for 22 h at which time [3H]thymidine incorporation was quantified. Significantly different from the value for T5 (*) or Tl (**) cells that did not receive TGF/3 (multiway analysis of variance, P < 0.05). Data are the mean ± SEM of two independent determinations, performed in duplicate. TABLE 3. Concentration of TGF/3 required for modulation of rat prostate cancer cell thymidine incorporation
Cell line Tl T5
TGF/3 required for half-maximum effect (ng/ml) Stimulation
Inhibition
0.36 ± 0.03 3.35 ± 0.15°
0.0027 ± 0.0009 0.0060 ± 0.0025
Data are the mean ± SEM for two independent determinations. Significantly different from the value for Tl cells, P < 0.05.
25-
T5 Cells
I 10
5-
0.02
0.1
0.2
1 2 bFCP (n^ml)
10
20
50
FIG. 4. TGFjS inhibits bFGF stimulation of Tl and T5 rat prostate cancer cell [3H]thymidine incorporation. Tl and T5 cells were cultured and growth arrested on 24-well plates as detailed in Fig. 3. Culture medium was then replaced with serum-free medium which either did not contain TGF/3 (S) or contained 0.01 (•) or 0.1 (•) ng TGF/3/ml. Cells were cultured for 3 h at 37 C, and wells then received the indicated amount of bFGF. After 22-h additional culture, thymidine incorporation was quantified. *, Significantly different from the value for cells that did not receive TGF/3 (multiway analysis of variance, P < 0.05). Data are the mean ± SEM for triplicate analyses.
0
Because we have shown that bFGF enhances rat prostate cancer cell thymidine incorporation (16), we sought to define the effect of TGF0 on bFGF mitogenicity in these cells. Titration analyses established that increasing medium bFGF content initially enhanced Tl or T5 cell thymidine incorporation and that maximum enhancement occurred when the medium bFGF content was 1-2 ng/ml (Fig. 4). Further increases in medium bFGF content caused progressive decrements in thymidine incorporation by T l or T5 cells (Fig. 4). Simultaneous inclusion of either 0.01 or 0.1 ng/ml TGF/3 significantly diminished the magnitude of bFGF stimulation of Tl or T5 cell thymidine incorporation. TGF/3-mediated inhibition of bFGF-enhanced prostate cancer cell thymidine incorporation was TGF/3 concentration dependent through that portion of the bFGF titration for which bFGF stimulated T l or T5 cell thymidine incorporation (Fig. 4). The concentration of bFGF required to achieve
TABLE 4. The effect of TGF/3 on the concentration of bFGF required for half-maximum stimulation of rat prostate cancer cell thymidine incorporation bFGF required for half-maximum stimulation (ng/ml) line
None
0.01 ng/ml TGF/3
0.10 ng/ml TGF/3
Tl T5
0.142 ± 0.019 0.083 ± 0.015
0.358 ± 0.093° 0.152 ± 0.002°
0.705 ±0.110 a ' 6 0.133 ± 0.011°
Data are the mean ± SEM for triplicate determinations. Significantly different from the value for similar cells which did not receive TGF0, P < 0.05. 6 Significantly different from the value for identically treated T5 cells, P < 0.05. 0
half-maximum stimulation of Tl cell thymidine incorporation significantly increased as the culture medium TGF/3 concentration was increased (Table 4). In contrast, although TGF/3 diminished the magnitude of the bFGF-mediated enhancement of T5 cell thymidine incorporation, the concentration of bFGF required to
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 19:32 For personal use only. No other uses without permission. . All rights reserved.
PROSTATE CANCER CELL TGF/3 RECEPTORS achieve half-maximum stimulation was independent of the medium TGF/3 content (Table 4). Androgens do not affect T5 prostate cancer cell TGF(3 receptor content We have shown that androgens inhibit in vitro proliferation of T5 prostate cancer cells and that the effects are mediated through functional androgen receptors (18, 19). To determine whether androgen effects may in part be mediated through changes in prostate cancer cell TGF/3 receptor content, we examined androgen modulation of T5 cell TGF/3 receptor content. These analyses demonstrated that physiological concentration of testosterone, which cause in vitro inhibition of T5 cell proliferation (18, 19), fail to affect T5 cell TGF/3 receptor content (Table 5). Our inability to detect an androgen effect on TGF/3 receptor content of these cells was not attributable to potential androgen-mediated changes in TGF/3 binding to its receptor, because we found that the Kd for TGF/3 binding to T5 cells is androgen concentration independent (Table 5). Discussion We show that AXC/SSh rat prostate cancer cells contain limited capacity, high affinity TGF/3-binding components which have mitogen binding specificity characteristic of TGF/3 receptors. T l cell TGF|3 receptor content is comparable to that reported (26) for cultured human prostate PC3 and DU145 cells which, respectively, contain 3,000 and 1,500 TGF0 receptors/cell. The TGF/3 receptor content of C3, Dl, or T5 cells is similar to that of osteoblasts (27, 28), NRK-49F (29, 30), or BALB/c 3T3 (29) cells which, respectively, contain 5,000-10,000, 17,000-19,000, or 16,000 TGF/3 receptors/ cell. In contrast, Swiss 3T3 cells contain 80,000 TGF,8 receptors/cell (31). The TGF/3 receptor content of normal rat ventral prostate membrane preparations is 50 fmol/mg DNA (6), approximately 300 sites/cell. Differences in the TGF/3 receptor content of normal rat ventral prostate (6) or rat prostate cancer cells may reflect differences in properties of the preparations analyzed. Osteoblast TGF/3 Kd has been reported to be 2.2 (26) or TABLE 5. AXC/SSh T5 rat prostate cancer cell TGF/3 receptor content is not modulated by testosterone Testosterone (M)
None 10"7 10"8 10"9
Sites/cell (% of control)
(pM)
Kd
100 118 ± 10 136 ± 19 108 ± 9
115 ± 15 96 ± 16 126 ± 1 110 ± 26
Data are the mean ± SEM for two independent determinations, performed in triplicate.
823
50-150 pM (27). In other cultured cells (25, 26, 29-31) or prostate membrane preparations (6) the Kd for TGF/3 ranges from 25-150 pM. These values are similar to the Kd of TGF/3 for rat prostate cancer cell receptors. The size of rat prostate cancer cell TGF/3 receptor (~280 kDa) is identical to that of normal rat ventral prostate (6) and the principal TGF/3 receptor of other mammalian cells (29, 31, 32). TGF/3 receptors ranging in size from 65-140 kDa have been described (27, 29, 31, 32). These species were not detected after chemical cross-linking of TGF/3 bound to rat prostate cancer cells (data not shown). TGF/3 is a bifunctional modulator of rat prostate cancer cell thymidine incorporation. Half-maximum inhibition of thymidine incorporation by T l or T5 cells requires 0.11-0.24 pM TGFjS, values at least 2 orders of magnitude less than that necessary to inhibit the function of numerous other cells. By example, half-maximum inhibition of soft agar colony formation (anchorage-independent growth) of multiple human carcinomas generally is achieved with 10-30 pM TGF/3 (33), whereas others suggest that half-maximum inhibition of colony formation of some human cells requires 320 pM TGF/3 (34). Similarly, enhancement of human osteosarcoma cell proliferation by either EGF or PDGF is inhibited by TGF/3 at a half-maximum concentration of 40 pM (35). TGF/3 inhibition of rat embryo fibroblast (36), NIH 3T3 (36), and NRK (37) colony formation, or primary rat hepatocyte thymidine incorporation (37) requires 1-15 pM TGF/3 for half-maximum effect. In contrast, halfmaximum inhibition of EGF-stimulated mink lung cell proliferation requires 300 pM TGF/3 (38). The concentration of TGF/3 required for half-maximum stimulation of Tl cell thymidine incorporation was approximately 9-fold greater than that needed to achieve the same effect in T5 cells. Because identical numbers of T l and T5 cells were growth arrested, and plating efficiency and survival of these cells are comparable, the difference in T l and T5 cell sensitivity to TGF/3-modulated thymidine incorporation would not appear to be attributable to cell density-mediated differences in TGF/3 sensitivity, as is characteristic of other cells (28, 39-41). Our studies do not unequivocally eliminate the possibility that the apparent greater sensitivity of Tl cells to TGF/3enhanced thymidine incorporation is due to elevated TGF(8 secretion. If this occurred, the quantity of TGF/3 required to achieve a comparable increase in T5 cell thymidine incorporation would be artificially elevated compared to that in Tl cells. This seems an unlikely explanation. If Tl cells produced more TGF/3 than T5 cells, the increased TGF/3 content of Tl cell cultures would cause Tl cells to appear more sensitive to TGF/3 inhibition of thymidine incorporation. This is not the case. Finally, the concentrations of TGF/3 required for
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 19:32 For personal use only. No other uses without permission. . All rights reserved.
824
PROSTATE CANCER CELL TGF/5 RECEPTORS
half-maximum stimulation of Tl and T5 cell thymidine incorporation are representative of the range of concentrations required for half-maximum increase in colony formation (33, 34,42), proliferation (28), or EGF receptor content (41) of other cells. Concentration-dependent bifunctional TGF/3 modulation of T l and T5 cell thymidine incorporation appears to be unique to these cells. Although a considerable literature establishes that TGF0 may either stimulate or inhibit cell function (27, 28, 33-42), both responses rarely affect the same function in a cell. In the known cases the response is modulated by other effectors. For example, TGF/3 inhibits EGF-mediated and potentiates PDGFmediated colony formation by Myc-1 cells (33). Similarly, the ability of TGF/3 to either decrease (40) or increase (41) NRK cell EGF receptor content is dependent on cell density; the former occurs only in sparse cultures, whereas the latter occurs only in confluent cultures. Moreover, TGF/3 enhancement of NRK cell EGF receptor content occurs as a consequence of elevated receptor synthesis (41), whereas the decrease in apparent EGF receptor content is probably due to changes in receptor Kd for EGF (40). The dose-response curve for bFGF modulation of Tl or T5 cell thymidine incorporation was bell shaped, demonstrating that high concentrations of bFGF depress thymidine incorporation by these cells. This finding essentially is identical to results we detailed in prior studies (16). TGF/3 inhibited bFGF enhancement of Tl and T5 cell thymidine incorporation, and the effect was TGF/3 concentration dependent through that range of the bFGF titration which produced enhancement of thymidine incorporation. The concentration of bFGF required to achieve half-maximum increase in T5 cell thymidine incorporation was independent of culture medium TGF/3 content and essentially identical to the value we previously reported (16). In contrast, the concentration of bFGF required to achieve a half-maximum increase in T l cell thymidine incorporation increased 5-fold as the culture medium TGF/3 concentration increased. These differences in the effect of TGF/3 on bFGF modulation of Tl and T5 cell thymidine incorporation imply potential differences in the mode of action by TGF/3 as a modulator of bFGF action in these cells. The fact that the concentration of bFGF necessary for half-maximum increase in T5 cell thymidine incorporation is independent of medium TGF(3 content suggests that TGF/3 affects T5 cell function at a point distal to bFGF interaction with T5 cell FGF receptors. This relationship is characteristic of TGF/? modulation of EGF-stimulated mink lung cell proliferation (38) or NRK cell EGF receptor content (41). In contrast, the fact that the concentration of bFGF required to achieve a half-maximum increase in Tl cell thymidine incorporation increases as the medium
Endo • 1990 Voll26«No2
TGF/3 content is increased implies that TGF/3 proximally affects bFGF function in T l cells. The half-maximum concentration of TGF/3 necessary to inhibit bFGF stimulation of bovine aortic arch (43, 44) or retinal capillary (45) endothelial cell proliferation is 1-30 pM. Thus, sensitivity of these cells to TGFjS inhibition of function is significantly less than that of rat prostate cancer cells. Because the bFGF concentration required for a half-maximum response was independent of medium TGF/3 content (43, 44), the effects of TGF/3 in these cells are distal to the bFGF receptor. Consequently, the mechanism of TGF/3 modulation of endothelial cell and T5 prostate cancer cell bFGF response may be similar. Results of our studies establish that rat prostate cancer cells contain functional TGF/3 receptors and imply the presence of functional bFGF receptors. The finding that TGFjS modulates bFGF effects establishes that mitogen modulation of prostate cancer cell function is multifactorial. The bifunctional TGF/3 modulation of prostate cancer cell DNA synthesis is unique and provides some insight into the potential complexity of mitogen modulation of prostate cancer cell proliferation. The mechanism by which these mitogens interact to affect cancer cell function is unknown. However, the differences we detail for T l and T5 cells implies that some interactive effects of mitogens may be cell line specific.
References 1. Griffin JE, Wilson JD 1980 The syndromes of androgen resistance. N Engl J Med 302:198 2. Huggins C, Hodges CV 1941 Studies in prostatic cancer. I. The effects of castration, of estrogen and of androgen injection on serum phosphatases in metastatic carcinoma of the prostate. Cancer Res 1:293 3. McKeehan WL, Adams PS, Rosser MP 1984 Direct mitogenic effects of insulin, epidermal growth factor, 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 4. Eaton CL, Davies P, Phillips MEA1988 Growth factor involvement and oncogene expression in prostatic tumours. J Steroid Biochem 30:341 5. Traish AM, Wotiz HH 1987 Prostatic epidermal growth factor receptors and their regulation by androgens. Endocrinology 121:1461 6. Kyprianou N, Isaacs JT 1988 Identification of a cellular receptor for transforming growth factor-/? in rat ventral prostate and its negative regulation by androgens. Endocrinology 123:2124 7. Davies P, Eaton CL 1989 Binding of epidermal growth factor by human normal, hypertrophic, and carcinomatous prostate. Prostate 14:123 8. Schuurmans ALG, Bolt J, Mulder E 1988 Androgens stimulate both growth rate and epidermal growth factor receptor activity of the human prostate tumor cell LNCaP. Prostate 12:55 9. Jacobs SC, Story MT, Sasse J, Lawson RK 1988 Characterization of growth factors derived from the rat ventral prostate. J Urol 139:1106 10. Maehama S, Li D, Nanri H, Leykam JF, Deuel TF 1986 Purification and partial characterization of prostate-derived growth factor. Proc Natl Acad Sci USA 83:8162 11. Ikeda T, Lioubin MN, Marquardt H 1987 Human transforming
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 19:32 For personal use only. No other uses without permission. . All rights reserved.
PROSTATE CANCER CELL TGF/3 RECEPTORS
12.
13. 14. 15.
16. 17. 18. 19. 20.
21. 22. 23. 24. 25. 26. 27.
28.
growth factor type /32: production by a prostatic adenocarcinoma cell line, purification, and initial characterization. Biochemistry 26:2406 Story MT, Esch F, Shimasaki S, Sasse J, Jacobs SC, Lawson RK 1987 Amino-terminal sequence of a large form of basic fibroblast growth factor isolated from human benign prostatic hyperplastic tissue. Biochem Biophys Res Commun 142:702 Story MT, Sasse J, Jacobs SC, Lawson RK 1987 Prostatic growth factor: purification and structural relationship to basic fibroblast growth factor. Biochemistry 26:3843 Mydlo JH, Michaeli J, Heston WDW, Fair WR 1988 Expression of basic fibroblast growth factor mRNA in benign prostatic hyperplasia and prostatic carcinoma. Prostate 13:241 Matuo Y, Nishi N, Matsui S-i, Sandberg AA, Isaacs JT, Wada F 1987 Heparin binding affinity of rat prostatic growth factor in normal and cancerous prostates: partial purification and characterization of rat prostatic growth factor in the Dunning tumor. Cancer Res 47:188 Shain SA, Koger JD 1989 Differences in responsiveness of clonally derived AXC/SSh rat prostate cancer cells to secreted or prototypic mitogens. Cancer Res 49:3898 Shain SA, Huot RI, Gorelic LS, Smith GC 1984 Biochemical and morphological characterization of clonal AXC rat prostate cancer cells. Cancer Res 44:2033 Huot RI, Shain SA 1986 Differential androgen modulation of AXC/ SSh rat prostate cancer cell proliferation in vitro and its antagonism by antiandrogen. Cancer Res 46:3775 Lin AL, Shain SA 1989 Androgen modulation of prostate cancer cell androgen receptor content is cell line specific. Mol Cell Endocrinol 63:75 Knabbe C, Lippman ME, Wakefield LM, Flanders KC, Kasid A, Derynck R, Dickson RB 1987 Evidence that transforming growth factor-0 is a hormonally regulated negative growth factor in human breast cancer cells. Cell 48:417 Scatchard G 1949 The attractions of proteins for small molecules and ions. Ann NY Acad Sci 51:660 Shain SA, Boesel RW 1978 Human prostate steroid hormone receptor quantitation. Current methodology and possible utility as a clinical discriminant in carcinoma. Invest Urol 16:169 Laemmli UK 1970 Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (Lond) 227:680 Nie NH, Hull CH, Jenkins JG, Steinbrenner K, Bent DH 1975 SPSS: Statistical Package for the Social Sciences, ed 2. McGrawHill, New York Sporn MB, Roberts AB, Wakefield LM, de Crombrugghe B 1987 Some recent advances in the chemistry and biology of transforming growth factor-beta. J Cell Biol 105:1039 Wilding G, Zugmeier G, Knabbe C, Flanders K, Gelmann E 1989 Differential effects of transforming growth factor /3 on human prostate cancer cells in vitro. Mol Cell Endocrinol 62:79 Centrella M, McCarthy TL, Canalis E 1988 Parathyroid hormone modulates transforming growth factor /? activity and binding in osteoblast-enriched cell cultures from fetal rat parietal bone. Proc Natl Acad Sci USA 85:5889 Robey PG, Young MF, Flanders KC, Roche NS, Kondaiah P, Reddi AH, Termine JD, Sporn MB, Roberts AB 1987 Osteoblasts synthesize and respond to transforming growth factor-type 0 (TGF-/3) in vitro. J Cell Biol 105:457
825
29. Massague J, Like B 1985 Cellular receptors for type /? transforming growth factor. Ligand binding and affinity labeling in human and rodent cell lines. J Biol Chem 260:2636 30. Frolik CA, Wakefield LM, Smith DM, Sporn MB 1984 Characterization of a membrane receptor for transforming growth factor-/? in normal rat kidney fibroblasts. J Biol Chem 259:10995 31. Fanger BO, Wakefield LM, Sporn MB 1986 Structure and properties of the cellular receptor for transforming growth factor type /3. Biochemistry 25:3083 32. Chiefetz S, Like B, Massague J 1986 Cellular distribution of type I and type II receptors for transforming growth factor-/9. J Biol Chem 261:9972 33. Roberts AB, Anzano MA, Wakefield LM, Roche NS, Stern DF, Sporn MB 1985 Type /? transforming growth factor: a bifunctional regulator of cellular growth. Proc Natl Acad Sci USA 82:119 34. Moses HL, Tucker RF, Leof EB, Coffey Jr RJ, Halper J, Shipley GD 1985 Type-/? transforming growth factor is a growth stimulator and a growth inhibitor. In: Feramisco J, Ozanne B, Stiles C (eds) Growth Factors and Transformation. Cold Spring Harbor Laboratory, Cold Spring Harbor, p. 65 35. Mioh H, Chen J-K 1989 Differential inhibitory effects of TGF-/3 on EGF, PDGF-, and HBFG-1-stimulated MG63 human osteosarcoma cell growth: possible involvement of growth factor interactions at the receptor and postreceptor levels. J Cell Physiol 139:509 36. Anzano MA, Roberts AB, Sporn MB 1986 Anchorage-independent growth of primary rat embryo cells is induced by platelet-derived growth factor and inhibited by type-beta transforming growth factor. J Cell Physiol 126:312 37. Nakamura T, Tomita Y, Hirai R, Yamaoka K, Kaji K, Ichihara A 1985 Inhibitory effect of transforming growth factor-/? on DNA synthesis of adult rat hepatocytes in primary culture. Biochem Biophys Research Commun 133:1042 38. Like B, Massague J 1986 The antiproliferative effect of type /3 transforming growth factor occurs at a level distal from receptors for growth-activating factors. J Biol Chem 261:13426 39. Centrella M, McCarthy TL, Canalis E 1987 Transforming growth factor /3 is a bifunctional regulator of replication and collagen synthesis in osteoblast-enriched cell cultures from fetal rat bone. J Biol Chem 262:2869 40. Massague J 1985 Transforming growth factor-/3 modulates the high-affinity receptors for epidermal growth factor and transforming growth factor-a. J Cell Biol 100:1508 41. Assoian RK, Frolik CA, Roberts AB, Miller DM, Sporn MB 1984 Transforming growth factor-/? controls receptor levels for epidermal growth factor in NRK fibroblasts. Cell 36:35 42. Tucker RF, Shipley GD, Moses HL, Holley RW 1984 Growth inhibitor from BSC-1 cells closely related to platelet type /? transforming growth factor. Science 226:705 43. Frater-Schroder M, Muller G, Birchmeier W, Bohlen P 1986 Transforming growth factor-beta inhibits endothelial cell proliferation. Biochem Biophys Res Commun 137:295 44. Baird A, Durkin T 1986 Inhibition of endothelial cell proliferation by type ^-transforming growth factor: interactions with acidic and basic fibroblast growth factors. Biochem Biophys Res Commun 138:476 45. Bensaid M, Malecaze F, Bayard F, Tauber JP 1989 Opposing effects of basic fibroblast growth factor and transforming growth factor-/? on the proliferation of cultured bovine retinal capillary endothelial (BREC) cells. Exp Eye Res 48:791
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 19:32 For personal use only. No other uses without permission. . All rights reserved.
