The Prostate 21:13>143 (1992)
Interaction of Epidermal Growth Factor and Transforming Growth Factor Beta in Human Prostatic Epithelial Cells in Culture Debra M. Sutkowski, Chau-Jye Fong, Julia A. Sensibar, Alfred W. Rademaker, Edward R. Sherwood, James M. Kozlowski, and Chung Lee Department of Urology and the Cancer Center, Northwestern University Medical School, Chicago, Illinois The present study was conducted to study the interaction between epidermal growth factor (EGF) and transforming growth factor$ (TGF-P) in benign human prostatic epithelial cells in culture. Primary cultures of human prostatic epithelial cells were grown in complete WAJC, which consisted of WAJC-404 medium and, in addition to other defined additives, EGF and bovine pituitary extract (BPE). Incomplete WAJC contained the same composition except EGF and BPE were deleted. TGF-P was added into media at concentrations of 0, 0.1, and 1.0 ng/ml. When cells were grown in complete WAJC, they proliferated rapidly. Cell proliferation was greatly suppressed when incomplete WAJC was used. Addition of TGF-P to these cultures caused a significant reduction in the final cell number when either complete WAJC or incomplete WAJC was used. In additional experiments, cells were prelabeled with 3H-thymidine for 72 hr prior to treatment with TGF-P. The percentage of radioactivity released into the medium at the end of a 6-day culture was used as an indication of the extent of cell death. Trypan blue exclusion test was also used to assess the extent of cell death. Addition of TGF-P into complete WAJC did not significantly affect the extent of cell death beyond what was considered as the result of normal cellular turnover. Addition of TGF-P into incomplete WAJC, however, caused a significant increase in the percent of cell death in the culture. These results demonstrated an interaction between EGF and TGF-P in proliferation and cell death in human prostatic epithelia in culture. In the presence of EGF alone in the culture medium, prostatic epithelial cells were stimulated to proliferate. The rate of proliferation was greatly diminished when EGF was deleted from the medium or when TGF-P was added in the presence of EGF. Finally, cell death was induced when TGF-P was added into the medium in the absence of EGF. 0 1992 Wiley-Liss, Inc.
Key words: cell proliferation, proliferation arrest, cell death
INTRODUCTION
Prostatic epithelial cells require androgen for proliferation and maintenance of normal function [ 1,2]. Removal of this androgenic support, under in vivo conditions,
Received for publication December 5 , 1991; accepted May 18, 1992. Address reprint requests to Dr. Chung Lee, Department of Urology, Northwestern University Medical School, 303 East Chicago Avenue, Chicago, IL 6061 1.
0 1992 Wiley-Liss, Inc.
134
Sutkowski et al.
results in a rapid rate of cell death [3]. However, findings of recent in vitro studies have suggested that androgen is not the direct regulatory agent. Rather, growth factors may play a more proximal role in regulating activities of these cells [4]. Among the growth factors, epidermal growth factor (EGF) and transforming growth factor beta (TGF-P) have been studied most extensively. Their effects on prostatic epithelial cells are briefly reviewed. EGF is mitogenic to prostatic epithelial cells in vitro [5,6].The level of EGF in the prostate is regulated by the androgenic status of the host. Castration in adult male mice results in prostatic involution, accompanied by a marked reduction in the amount of EGF in the prostate. Conversely, replacement of testosterone in these animals stimulates prostatic growth and restores the tissue levels of EGF [7]. On the other hand, TGF-/3 is known to be a growth inhibitor for a variety of epithelial cells [8,9]. Local administration of TGF-P could induce epithelial cell death in the rat prostate [ 101. Furthermore, the level of mRNA for TGF-P is significantly elevated in the rat prostate during castration-induced involution [ 111. On the basis of the above consideration, it is apparent that EGF and TGF-P are playing opposite roles in prostatic epithelial cells. The former has a stimulatory effect, while the latter has an inhibitory effect. If these two growth factors are indeed important in the regulation of prostatic proliferation and turnover, they must interact with one another in a manner to achieve prostatic homeostasis. The present study was conducted in an attempt to determine the possibility of whether such an interaction exists between EGF and TGF-P in prostatic epithelial cells. Benign human prostatic epithelial cells were used as a target. Our results indicated that prostatic epithelial cells proliferated in the presence of EGF alone but that proliferation was suppressed when TGF-P was administered in the presence of EGF. Further studies showed that these cells underwent cell death when they were exposed to TGF-P in the absence of EGF. MATERIALS AND METHODS Human Prostatic Epithelial Cell Culture
Human prostatic epithelial cells were isolated from prostatic tissue obtained from patients undergoing open prostatectomy for treatment of bladder neck obstruction secondary to benign prostatic hyperplasia [12,13]. The diagnosis was confirmed by review of histological sections of representative tissue specimens. Briefly, surgical specimens of the prostate were mechanically dissociated into small pieces of 1 mm3 by means of surgical scalpels. Tissue pieces were further dissociated enzymatically in RPMI-1640 medium containing 10%fetal bovine serum, DNAse (100 pg/ml, Sigma Chemical Co., St. Louis, MO), and type I collagenase (200 Ulml) (Sigma Chemical Co.) for a period of 16 hr. The dissociated stromal and epithelial cells were separated using discontinuous percoll gradient centrifugation. Epithelial cells were selectively cultured in complete WAJC, which consisted of WAJC-404 medium [5,6] and the following additives: ITS (insulin, 5 pg/ml; transferrin, 5 pg/ml; selenious acid, 5 ng/ml) (Collaborative Research); EGF (10 ng/ml) (Collaborative Research); bovine pituitary extract (BPE, 30 pg/ml) (Collaborative Research); choleratoxin (10 ng/ml) (Sigma Chemical Co); prolactin (2.3 pg/ml) (Sigma Chemical Co.); polyvinylpyrrolidone (2 mg/ml) (Sigma Chemical Co.); penicillin, 100 U/ml; and streptomycin, 100 pg/ml. Cells were cultivated in T75 flasks under routine conditions. Results of
EGF and TGF-f3 in Prostatic Epithelium
135
our previous studies show that purified epithelial cells can be obtained through the employment of these procedures [ 12,131. Experimental Design
After a period of 10 days of culture in flasks following the initial harvest from the percoll gradient, prostatic epithelial cells were dissociated from the flask surface by brief treatment of cells with trypsin (0.25%), washed, and reseeded in 24-well plates at a rate of 3.0 X lo4 cells per well. The following four experiments were conducted. Experiment 1. This experiment was conducted to test the mitogenic potential of EGF and TGF-P. Purified TGF-P, was purchased from Collaborative Research. TGF-P, at 0, 0.1, or 1.O ng/ml, was added into either complete WAJC or incomplete WAJC. The latter medium was identical to complete WAJC except that EGF and BPE were deleted from the formulation. Media were changed on days 2 and 4. Cells were harvested on day 6 and counted in a Coulter counter. The number of adherent cells present at the end of the experiment was used as an indication of the mitogenic potential of the medium. Experiment 2. This experiment was carried out to test the cytotoxic potential of various media. The above experiment was repeated except that cells were prelabeled with 3H-thymidine (5 pCi/ml, Amersham) for 72 hr prior to seeding in 24-well plates. Media were changed on days 2 and 4. Cells were harvested on day 6, and the cumulative amount of radioactivity released into cultured media was determined by liquid scintillation counting. The percent radioactivity released was calculated as Total dpm - dpm remaining in cells on day 6 total dpm = percentage radioactivity released,
x 100
where total dpm represented the amount of radioactivity present in cells as well as that released into media during 6 days of culture. The percentage of radioactivity released by cells cultured in complete WAJC was used as an indication of the normal turnover rate. A significant increase in the percentage of radioactivity released in any treatment beyond what was represented by the normal turnover rate was considered as having a positive cytotoxic effect. Experiment 3. This experiment was designed in an attempt to delineate the individual role of EGF and BPE. Both the mitogenesis assay and cell lysis assay were conducted. The above two experiments were expanded so that the following groups of medium were included; complete WAJC, complete WAJC with EGF deleted, complete WAJC with BPE deleted, and incomplete WAJC. Media were changed on days 2 and 4. On day 6, the number of cells was counted and the amount of radioactivity released into the medium was calculated as the percent of total radioactivity. Experiment 4. This experiment was designed to confirm the cytotoxic potential of the various media. TGF-P was added into either complete WAJC or incomplete WAJC at 0 or 1 ng/ml. Media was changed on days 2 and 4. Cells were harvested on days 2, 4, and 6 and stained with trypan blue to determine viability. The percent of nonviable cells was calculated as follows:
136
Sutkowski et al. 250000
200000
I
L
Q)
P 150000
E
3
z
-
I
100000
Q)
0 50000
0
I
0
1
0.1
0
0.1
1.0
T G F- beta, ng/ml Fig. 1 . Effect of TGF-P on proliferation of human prostate epithelial cells in media containing EGF and BPE ([II,complete WAJC) or in media without EGF or BPE (m, incomplete WAJC). Primary cultures were grown in various media for 6 days. Cell number at the end of 6-day period was counted with a Coulter counter. Vertical bars denote standard error with N = 4.
Trypan blue positive cells total number of cells
x 100,
where total cell number represented both trypan blue positive cells, and cells excluding the dye. Statistics All numerical data were expressed as mean ?SE, calculated on the basis of quadruplicate observations, except for experiment 4 which had triplicate observations. Differences among treatment groups were determined by analysis of variance test followed by Duncan’s multiple-range test [14]. A value of P
Interaction of Epidermal Growth Factor and Transforming Growth Factor Beta in Human Prostatic Epithelial Cells in Culture Debra M. Sutkowski, Chau-Jye Fong, Julia A. Sensibar, Alfred W. Rademaker, Edward R. Sherwood, James M. Kozlowski, and Chung Lee Department of Urology and the Cancer Center, Northwestern University Medical School, Chicago, Illinois The present study was conducted to study the interaction between epidermal growth factor (EGF) and transforming growth factor$ (TGF-P) in benign human prostatic epithelial cells in culture. Primary cultures of human prostatic epithelial cells were grown in complete WAJC, which consisted of WAJC-404 medium and, in addition to other defined additives, EGF and bovine pituitary extract (BPE). Incomplete WAJC contained the same composition except EGF and BPE were deleted. TGF-P was added into media at concentrations of 0, 0.1, and 1.0 ng/ml. When cells were grown in complete WAJC, they proliferated rapidly. Cell proliferation was greatly suppressed when incomplete WAJC was used. Addition of TGF-P to these cultures caused a significant reduction in the final cell number when either complete WAJC or incomplete WAJC was used. In additional experiments, cells were prelabeled with 3H-thymidine for 72 hr prior to treatment with TGF-P. The percentage of radioactivity released into the medium at the end of a 6-day culture was used as an indication of the extent of cell death. Trypan blue exclusion test was also used to assess the extent of cell death. Addition of TGF-P into complete WAJC did not significantly affect the extent of cell death beyond what was considered as the result of normal cellular turnover. Addition of TGF-P into incomplete WAJC, however, caused a significant increase in the percent of cell death in the culture. These results demonstrated an interaction between EGF and TGF-P in proliferation and cell death in human prostatic epithelia in culture. In the presence of EGF alone in the culture medium, prostatic epithelial cells were stimulated to proliferate. The rate of proliferation was greatly diminished when EGF was deleted from the medium or when TGF-P was added in the presence of EGF. Finally, cell death was induced when TGF-P was added into the medium in the absence of EGF. 0 1992 Wiley-Liss, Inc.
Key words: cell proliferation, proliferation arrest, cell death
INTRODUCTION
Prostatic epithelial cells require androgen for proliferation and maintenance of normal function [ 1,2]. Removal of this androgenic support, under in vivo conditions,
Received for publication December 5 , 1991; accepted May 18, 1992. Address reprint requests to Dr. Chung Lee, Department of Urology, Northwestern University Medical School, 303 East Chicago Avenue, Chicago, IL 6061 1.
0 1992 Wiley-Liss, Inc.
134
Sutkowski et al.
results in a rapid rate of cell death [3]. However, findings of recent in vitro studies have suggested that androgen is not the direct regulatory agent. Rather, growth factors may play a more proximal role in regulating activities of these cells [4]. Among the growth factors, epidermal growth factor (EGF) and transforming growth factor beta (TGF-P) have been studied most extensively. Their effects on prostatic epithelial cells are briefly reviewed. EGF is mitogenic to prostatic epithelial cells in vitro [5,6].The level of EGF in the prostate is regulated by the androgenic status of the host. Castration in adult male mice results in prostatic involution, accompanied by a marked reduction in the amount of EGF in the prostate. Conversely, replacement of testosterone in these animals stimulates prostatic growth and restores the tissue levels of EGF [7]. On the other hand, TGF-/3 is known to be a growth inhibitor for a variety of epithelial cells [8,9]. Local administration of TGF-P could induce epithelial cell death in the rat prostate [ 101. Furthermore, the level of mRNA for TGF-P is significantly elevated in the rat prostate during castration-induced involution [ 111. On the basis of the above consideration, it is apparent that EGF and TGF-P are playing opposite roles in prostatic epithelial cells. The former has a stimulatory effect, while the latter has an inhibitory effect. If these two growth factors are indeed important in the regulation of prostatic proliferation and turnover, they must interact with one another in a manner to achieve prostatic homeostasis. The present study was conducted in an attempt to determine the possibility of whether such an interaction exists between EGF and TGF-P in prostatic epithelial cells. Benign human prostatic epithelial cells were used as a target. Our results indicated that prostatic epithelial cells proliferated in the presence of EGF alone but that proliferation was suppressed when TGF-P was administered in the presence of EGF. Further studies showed that these cells underwent cell death when they were exposed to TGF-P in the absence of EGF. MATERIALS AND METHODS Human Prostatic Epithelial Cell Culture
Human prostatic epithelial cells were isolated from prostatic tissue obtained from patients undergoing open prostatectomy for treatment of bladder neck obstruction secondary to benign prostatic hyperplasia [12,13]. The diagnosis was confirmed by review of histological sections of representative tissue specimens. Briefly, surgical specimens of the prostate were mechanically dissociated into small pieces of 1 mm3 by means of surgical scalpels. Tissue pieces were further dissociated enzymatically in RPMI-1640 medium containing 10%fetal bovine serum, DNAse (100 pg/ml, Sigma Chemical Co., St. Louis, MO), and type I collagenase (200 Ulml) (Sigma Chemical Co.) for a period of 16 hr. The dissociated stromal and epithelial cells were separated using discontinuous percoll gradient centrifugation. Epithelial cells were selectively cultured in complete WAJC, which consisted of WAJC-404 medium [5,6] and the following additives: ITS (insulin, 5 pg/ml; transferrin, 5 pg/ml; selenious acid, 5 ng/ml) (Collaborative Research); EGF (10 ng/ml) (Collaborative Research); bovine pituitary extract (BPE, 30 pg/ml) (Collaborative Research); choleratoxin (10 ng/ml) (Sigma Chemical Co); prolactin (2.3 pg/ml) (Sigma Chemical Co.); polyvinylpyrrolidone (2 mg/ml) (Sigma Chemical Co.); penicillin, 100 U/ml; and streptomycin, 100 pg/ml. Cells were cultivated in T75 flasks under routine conditions. Results of
EGF and TGF-f3 in Prostatic Epithelium
135
our previous studies show that purified epithelial cells can be obtained through the employment of these procedures [ 12,131. Experimental Design
After a period of 10 days of culture in flasks following the initial harvest from the percoll gradient, prostatic epithelial cells were dissociated from the flask surface by brief treatment of cells with trypsin (0.25%), washed, and reseeded in 24-well plates at a rate of 3.0 X lo4 cells per well. The following four experiments were conducted. Experiment 1. This experiment was conducted to test the mitogenic potential of EGF and TGF-P. Purified TGF-P, was purchased from Collaborative Research. TGF-P, at 0, 0.1, or 1.O ng/ml, was added into either complete WAJC or incomplete WAJC. The latter medium was identical to complete WAJC except that EGF and BPE were deleted from the formulation. Media were changed on days 2 and 4. Cells were harvested on day 6 and counted in a Coulter counter. The number of adherent cells present at the end of the experiment was used as an indication of the mitogenic potential of the medium. Experiment 2. This experiment was carried out to test the cytotoxic potential of various media. The above experiment was repeated except that cells were prelabeled with 3H-thymidine (5 pCi/ml, Amersham) for 72 hr prior to seeding in 24-well plates. Media were changed on days 2 and 4. Cells were harvested on day 6, and the cumulative amount of radioactivity released into cultured media was determined by liquid scintillation counting. The percent radioactivity released was calculated as Total dpm - dpm remaining in cells on day 6 total dpm = percentage radioactivity released,
x 100
where total dpm represented the amount of radioactivity present in cells as well as that released into media during 6 days of culture. The percentage of radioactivity released by cells cultured in complete WAJC was used as an indication of the normal turnover rate. A significant increase in the percentage of radioactivity released in any treatment beyond what was represented by the normal turnover rate was considered as having a positive cytotoxic effect. Experiment 3. This experiment was designed in an attempt to delineate the individual role of EGF and BPE. Both the mitogenesis assay and cell lysis assay were conducted. The above two experiments were expanded so that the following groups of medium were included; complete WAJC, complete WAJC with EGF deleted, complete WAJC with BPE deleted, and incomplete WAJC. Media were changed on days 2 and 4. On day 6, the number of cells was counted and the amount of radioactivity released into the medium was calculated as the percent of total radioactivity. Experiment 4. This experiment was designed to confirm the cytotoxic potential of the various media. TGF-P was added into either complete WAJC or incomplete WAJC at 0 or 1 ng/ml. Media was changed on days 2 and 4. Cells were harvested on days 2, 4, and 6 and stained with trypan blue to determine viability. The percent of nonviable cells was calculated as follows:
136
Sutkowski et al. 250000
200000
I
L
Q)
P 150000
E
3
z
-
I
100000
Q)
0 50000
0
I
0
1
0.1
0
0.1
1.0
T G F- beta, ng/ml Fig. 1 . Effect of TGF-P on proliferation of human prostate epithelial cells in media containing EGF and BPE ([II,complete WAJC) or in media without EGF or BPE (m, incomplete WAJC). Primary cultures were grown in various media for 6 days. Cell number at the end of 6-day period was counted with a Coulter counter. Vertical bars denote standard error with N = 4.
Trypan blue positive cells total number of cells
x 100,
where total cell number represented both trypan blue positive cells, and cells excluding the dye. Statistics All numerical data were expressed as mean ?SE, calculated on the basis of quadruplicate observations, except for experiment 4 which had triplicate observations. Differences among treatment groups were determined by analysis of variance test followed by Duncan’s multiple-range test [14]. A value of P
