JOURNAL OF CELLULAR PHYSIOLOGY 150:214219 (1992)
Regulation of Prostaglandin H Synthase Activity in Dog Urothelial Cells TERRY V. ZENSER,* DORl J. THOMAS, AMMlNl K. JACOB,AND BERNARD B. DAVIS V A Medical Center and Departments of Biochemistry and lnternal Medicine, St. Louis University School of Medicine, S t . Louis, Missouri 63 125 TPA regulation of prostaglandin H synthase activity in primary and subcultured dog urothelial cells was investigated. Previous studies have demonstrated an early (0-2 hr) increase in PGE, synthesis mediated by TPA which is dependent upon release of endogenous arachidonic acid by a phospholipase-mediated pathway. in this study, prostaglandin H synthase activity was assessed directly with microsomes and indirectly after addition of exogenous arachidonic acid at a maximum effective concentration (100 pM) to media. PGE, synthesis, measured by radioimmunoassay, served as an index of prostaglandin H synthase activity. After a 24-hr incubation with 0.1 pM TPA or 1 .O pM A23187, arachidonic acid elicited significantly more PGE, synthesis in agonist-treated cells than it did in control cells in primary culture. Microsomes from 24-hr TPA-treated cells exhibited significantly more prostaglandin H synthase activity than did those from control cells. In addition, the PGE, content of overnight media was approximately lo-fold greater in TPA-treated cells than in control cells. The late (24 hr) response was more sensitive to lower concentrations of TPA than was the earlier (0-2 hr) response. TPA at 0.1 pM was a maximum effective dose for both responses. The 24-hr response was blocked by cycloheximide and staurosporine, inhibitors of protein synthesis and protein kinase C, respectively. Pretreatment of cells with aspirin, an irreversible inhibitor of prostaglandin H synthase, prior to addition of TPA did not prevent the late TPA-mediated increase in PGE, synthesis. Subcultured cells exhibited both an early and a late TPA response. Only the early response was inhibited by aspirin pretreatment. Results suggest that the late response with TPA is caused by de novo synthesis of prostaglandin H synthase. Thus, primary and subcultured dog urothelial cells possess two distinct mechanisms for regulating signal transduction by arachidonic acid metabolism. This study provides a basis for assessing these mechanisms of signal transduction in urothelial cell lines and transformed cells
Hormones and growth factors transmit signals by second messengers which maintain cell growth, proliferation, and differentiation. Arachidonic acid metabolites can serve a s second or subsequent messengers in cells (Samuelsson et al., 1978; Smith 1989). Products of arachidonic acid metabolism include prostaglandins, thromboxanes, and leukotrienes. Because these compounds are effectively cleared by major organs (i.e., lungs and liver) and only very small amounts are found in the circulation, they are thought to function a s “local” hormones. Numerous studies indicate a role for arachidonic acid metabolites in carcinogenesis (for review see Powles et al., 1982; Thaler-Dao et al., 1984; Fischer and Slaga, 1985; Marnett, 1985). Because most malignancies are carcinomas derived from epithelial cells, the regulation of arachidonic acid metabolism by epithelial cells is important to understand. Significant contributions in the regulation of arachidonic acid metabolism have been made with cultured cells of nonepithelial origin. However, because results are often not compared with primary cultures of corresponding cells, the effect of acclimation to culture conditions is unknown. In addition, regulation of events in nonepi0 1992 WILEY-LISS, INC
thelial and epithelial cells may differ. For these reasons, studies in primary and subcultured urothelial cells have been initiated. This study provides a basis for assessing similar mechanisms of signal transduction in urothelial cell lines and transformed cells. Regulation of arachidonic acid metabolism can be separated into two distinct areas (Samuelsson et al., 1978; Smith, 1989). The first relates to the availability of arachidonic acid t i e . , substrate). Arachidonic acid is not found free within the cell, but is stored almost completely as esters of the sn-2 position of the glycerol backbone of cell membrane phospholipids. Phospholipases are necessary for arachidonic acid release and availability. The second area relates to metabolism of free unesterified arachidonic acid. Prostaglandin H synthase and lipoxygenases are the two enzymes most notable in this area. Cytochrome P-450s can also me-
Received January 23,1991; accepted August 22,1991. “To whom reprint requestsicorrespondence should be addressed.
REGULATION OF PROSTAGLANDIN H SYNTHASE ACTIVITY
tabolize arachidonic acid (Fitzpatrick and Murphy, 1989). De novo synthesis of prostaglandin H synthase can be induced by interleukin 1, epidermal growth factor (EGF), TPA, transforming growth factor beta, and interleukin 2 (Beaudry et al., 1985; Pash and Bailey, 1988; Frasier-Scott et al., 1988; Raz et al., 1989; Coyne et al., 1990). In studies of arachidonic acid metabolism in human and dog urothelial cells, similar results have been observed with agonists that include TPA, A23187, EGF, and bradykinin (Danon et al., 1986; Zenser et al., 1988, 1990; Zenser and Davis, 1988; Wong et al., 1989). An early agonist response (5-120 min) was observed (Zenser e t al., 1988, 1990; Wong et al., 1989). This early response was characterized as phospholipase-mediated because increased release of endogenous arachidonic acid occurred. With TPA, de novo synthesis of prostaglandin H synthase did not occur during the early urothelial response because combinations of maximally stimulating concentrations of TPA and arachidonic acid were not different from those observed with arachidonic acid alone (Zenser et al., 1988; Wong e t al., 1989). In addition, pretreatment with aspirin, a n irreversible inhibitor of synthase, completely inhibited TPA increases in PGE, synthesis (Zenser et al., 1990). The early response was also inhibited by cycloheximide and staurosporine, indicating involvement of protein synthesis and protein kinase c . The current study was designed to determine whether or not agonist-mediated signal transduction causes de novo synthesis of prostaglandin H synthase and, if so, by what mechanism. Prostaglandin H synthase activity was assessed directly with microsomes and indirectly after addition of exogenous arachidonic acid a t a maximum effective concentration (100 pM) to media. PGE,, the major arachidonic acid metabolite (Danon et al., 1986; Zenser et al., 1990), served a s a n index of prostaglandin H synthase activity. TPA and A23187 were examined in the current study because their mechanism of action is well characterized and their early response was previously assessed in urothelial cells (Zenser et al., 1988, 1990; Wong et al., 1989). TPA can substitute for diacylglycerol and stimulate protein kinase C (for review see Nishizuka, 1986; Castagna, 1987; Blumberg, 1988).A23187 is a calcium ionophore. Protein kinase C and/or calcium are involved in agonist-mediated arachidonic acid metabolism in many tissues. Results demonstrate two distinct mechanisms for regulating signal transduction by arachidonic acid metabolism in primary and subcultured urothelial cells.
MATERIALS AND METHODS Preparation of cell cultures Urothelial cells from dog urinary bladder were cultured using a modification of a method previously described for human urothelial cells (Reznikoff et al., 1983).Bladders were obtained from anesthetized mongrel dogs and placed into cold Hank's balanced salt solution (HBSS) (Wong e t al., 1989). Using aseptic techniques, the bladder was opened longitudinally and the mucosal surface was exposed. The urothelial cells were gently scraped from the underlying stroma using a scalpel blade. The detached cells were centrifuged at 1,500
215
rpm for 10 min at 4"C, and the cell pellet was resuspended in a small volume of media. Cells in 3 ml of media containing 1%FCS were added to 35 mm plastic tissue culture plates coated with 0.2% gelatin (Fisher Scientific Co., St. Louis, MO), and incubated at 37°C in 5% C02-95% humidified air. Media were changed biweekly and consisted of Hams Nutrient mixture F12 (GIBCO, Grand Island, NY) supplemented with 1% fetal calf serum and the following (final concentration in parentheses): insulin (10 pglml), hydrocortisone (1pg/ ml), transferrin (5 pgiml), and amphotericin B (2.5 pg/ ml), which were obtained from Sigma Chemical Co., St. Louis, MO; nonessential amino acids (0.1 mM), L-glutamine (2.0 mM), streptomycin (100 pg/ml), and penicillin (100 units/ml), which were purchased from GIBCO; and dextrose (2.7 mg/ml) obtained from Fisher Scientific. This medium was developed for human uroepithelial cells (Reznikoff et al., 1983, 1987). For subculturing, gelatin-coated T-75 flasks were used for primary cultures. After 10 days of primary culture, cells were dispersed for serial passage with 0.01% EDTA (Sigma Chemical Co.) and 0.25% trypsin (GIBCO) dissolved in HBSS (GIBCO) and added to gelatin-coated 35 mm plates as described above. Experiments were conducted with 80-90% confluent cultures (3-5 x lo5 cells per 35 mm plate). Growth media were aspirated and cells washed twice with 1ml of HBSS. Serum-free alpha-MEM was then added with or without test agents as indicated in the Results. Cells were incubated at 37°C in 5% C02-95%air. Test agents were arachidonic acid purchased from Nu-Chek Prep, Inc., Elysian, MN; calcium ionophore A23187 and staurosporine from CalBiochem, La Jolla, CA; and cycloheximide, aspirin, and TPA from Sigma Chemical Co., St. Louis, MO. Each agonist was used at a maximally effective concentration previously determined (Wong et al., 1989; Zenser e t al., 1990).
Radioimmunoassay of prostaglandin E, The PGE, content of the media was measured by double-antibody radioimmunoassay (Zenser and Davis, 1978). Rabbit antiserum to PGE, was obtained from Regis Chemical Co., Morton Grove, IL. Tritium-labeled PGE, was purchased from Du Pont NEN Research Products, Boston, MA; and goat antiserum to rabbit gamma globulins was purchased from Antibodies Inc., Davis, CA. Each condition was evaluated in two or three separate experiments. In each experiment, conditions were evaluated in a t least triplicate plates. The medium from each plate was analyzed in duplicate, and the value of duplicate determinations averaged and considered as a n N of one. Data were expressed as mean SEM of ng PGE, per lo5 cells. Differences between data were evaluated statistically by the Student's t-test for unpaired observations.
*
Prostaglandin H synthase assay Following a 24-hr incubation of cells in the absence and presence of TPA, media were saved for PGE, radioimmunoassay and microsomes were prepared for direct assessment of synthase activity (Brown et al., 1980). All subsequent procedures were performed a t 4°C. Cells were scraped into tubes, centrifuged at 2,000 rpm for 10 min to pellet cells, and were resuspended in buffer and
ZENSER ET AL.
216 50
40 h
m
T
0B A S A L ARACHIDONIC
PGE, PIIS ACTIVITY
E
2
1200
\
400
2
0
e 900
*" 30 0
?
k
r(
\
300
v
N 3
20
w 0 a
W
a >
v
Regulation of Prostaglandin H Synthase Activity in Dog Urothelial Cells TERRY V. ZENSER,* DORl J. THOMAS, AMMlNl K. JACOB,AND BERNARD B. DAVIS V A Medical Center and Departments of Biochemistry and lnternal Medicine, St. Louis University School of Medicine, S t . Louis, Missouri 63 125 TPA regulation of prostaglandin H synthase activity in primary and subcultured dog urothelial cells was investigated. Previous studies have demonstrated an early (0-2 hr) increase in PGE, synthesis mediated by TPA which is dependent upon release of endogenous arachidonic acid by a phospholipase-mediated pathway. in this study, prostaglandin H synthase activity was assessed directly with microsomes and indirectly after addition of exogenous arachidonic acid at a maximum effective concentration (100 pM) to media. PGE, synthesis, measured by radioimmunoassay, served as an index of prostaglandin H synthase activity. After a 24-hr incubation with 0.1 pM TPA or 1 .O pM A23187, arachidonic acid elicited significantly more PGE, synthesis in agonist-treated cells than it did in control cells in primary culture. Microsomes from 24-hr TPA-treated cells exhibited significantly more prostaglandin H synthase activity than did those from control cells. In addition, the PGE, content of overnight media was approximately lo-fold greater in TPA-treated cells than in control cells. The late (24 hr) response was more sensitive to lower concentrations of TPA than was the earlier (0-2 hr) response. TPA at 0.1 pM was a maximum effective dose for both responses. The 24-hr response was blocked by cycloheximide and staurosporine, inhibitors of protein synthesis and protein kinase C, respectively. Pretreatment of cells with aspirin, an irreversible inhibitor of prostaglandin H synthase, prior to addition of TPA did not prevent the late TPA-mediated increase in PGE, synthesis. Subcultured cells exhibited both an early and a late TPA response. Only the early response was inhibited by aspirin pretreatment. Results suggest that the late response with TPA is caused by de novo synthesis of prostaglandin H synthase. Thus, primary and subcultured dog urothelial cells possess two distinct mechanisms for regulating signal transduction by arachidonic acid metabolism. This study provides a basis for assessing these mechanisms of signal transduction in urothelial cell lines and transformed cells
Hormones and growth factors transmit signals by second messengers which maintain cell growth, proliferation, and differentiation. Arachidonic acid metabolites can serve a s second or subsequent messengers in cells (Samuelsson et al., 1978; Smith 1989). Products of arachidonic acid metabolism include prostaglandins, thromboxanes, and leukotrienes. Because these compounds are effectively cleared by major organs (i.e., lungs and liver) and only very small amounts are found in the circulation, they are thought to function a s “local” hormones. Numerous studies indicate a role for arachidonic acid metabolites in carcinogenesis (for review see Powles et al., 1982; Thaler-Dao et al., 1984; Fischer and Slaga, 1985; Marnett, 1985). Because most malignancies are carcinomas derived from epithelial cells, the regulation of arachidonic acid metabolism by epithelial cells is important to understand. Significant contributions in the regulation of arachidonic acid metabolism have been made with cultured cells of nonepithelial origin. However, because results are often not compared with primary cultures of corresponding cells, the effect of acclimation to culture conditions is unknown. In addition, regulation of events in nonepi0 1992 WILEY-LISS, INC
thelial and epithelial cells may differ. For these reasons, studies in primary and subcultured urothelial cells have been initiated. This study provides a basis for assessing similar mechanisms of signal transduction in urothelial cell lines and transformed cells. Regulation of arachidonic acid metabolism can be separated into two distinct areas (Samuelsson et al., 1978; Smith, 1989). The first relates to the availability of arachidonic acid t i e . , substrate). Arachidonic acid is not found free within the cell, but is stored almost completely as esters of the sn-2 position of the glycerol backbone of cell membrane phospholipids. Phospholipases are necessary for arachidonic acid release and availability. The second area relates to metabolism of free unesterified arachidonic acid. Prostaglandin H synthase and lipoxygenases are the two enzymes most notable in this area. Cytochrome P-450s can also me-
Received January 23,1991; accepted August 22,1991. “To whom reprint requestsicorrespondence should be addressed.
REGULATION OF PROSTAGLANDIN H SYNTHASE ACTIVITY
tabolize arachidonic acid (Fitzpatrick and Murphy, 1989). De novo synthesis of prostaglandin H synthase can be induced by interleukin 1, epidermal growth factor (EGF), TPA, transforming growth factor beta, and interleukin 2 (Beaudry et al., 1985; Pash and Bailey, 1988; Frasier-Scott et al., 1988; Raz et al., 1989; Coyne et al., 1990). In studies of arachidonic acid metabolism in human and dog urothelial cells, similar results have been observed with agonists that include TPA, A23187, EGF, and bradykinin (Danon et al., 1986; Zenser et al., 1988, 1990; Zenser and Davis, 1988; Wong et al., 1989). An early agonist response (5-120 min) was observed (Zenser e t al., 1988, 1990; Wong et al., 1989). This early response was characterized as phospholipase-mediated because increased release of endogenous arachidonic acid occurred. With TPA, de novo synthesis of prostaglandin H synthase did not occur during the early urothelial response because combinations of maximally stimulating concentrations of TPA and arachidonic acid were not different from those observed with arachidonic acid alone (Zenser et al., 1988; Wong e t al., 1989). In addition, pretreatment with aspirin, a n irreversible inhibitor of synthase, completely inhibited TPA increases in PGE, synthesis (Zenser et al., 1990). The early response was also inhibited by cycloheximide and staurosporine, indicating involvement of protein synthesis and protein kinase c . The current study was designed to determine whether or not agonist-mediated signal transduction causes de novo synthesis of prostaglandin H synthase and, if so, by what mechanism. Prostaglandin H synthase activity was assessed directly with microsomes and indirectly after addition of exogenous arachidonic acid a t a maximum effective concentration (100 pM) to media. PGE,, the major arachidonic acid metabolite (Danon et al., 1986; Zenser et al., 1990), served a s a n index of prostaglandin H synthase activity. TPA and A23187 were examined in the current study because their mechanism of action is well characterized and their early response was previously assessed in urothelial cells (Zenser et al., 1988, 1990; Wong et al., 1989). TPA can substitute for diacylglycerol and stimulate protein kinase C (for review see Nishizuka, 1986; Castagna, 1987; Blumberg, 1988).A23187 is a calcium ionophore. Protein kinase C and/or calcium are involved in agonist-mediated arachidonic acid metabolism in many tissues. Results demonstrate two distinct mechanisms for regulating signal transduction by arachidonic acid metabolism in primary and subcultured urothelial cells.
MATERIALS AND METHODS Preparation of cell cultures Urothelial cells from dog urinary bladder were cultured using a modification of a method previously described for human urothelial cells (Reznikoff et al., 1983).Bladders were obtained from anesthetized mongrel dogs and placed into cold Hank's balanced salt solution (HBSS) (Wong e t al., 1989). Using aseptic techniques, the bladder was opened longitudinally and the mucosal surface was exposed. The urothelial cells were gently scraped from the underlying stroma using a scalpel blade. The detached cells were centrifuged at 1,500
215
rpm for 10 min at 4"C, and the cell pellet was resuspended in a small volume of media. Cells in 3 ml of media containing 1%FCS were added to 35 mm plastic tissue culture plates coated with 0.2% gelatin (Fisher Scientific Co., St. Louis, MO), and incubated at 37°C in 5% C02-95% humidified air. Media were changed biweekly and consisted of Hams Nutrient mixture F12 (GIBCO, Grand Island, NY) supplemented with 1% fetal calf serum and the following (final concentration in parentheses): insulin (10 pglml), hydrocortisone (1pg/ ml), transferrin (5 pgiml), and amphotericin B (2.5 pg/ ml), which were obtained from Sigma Chemical Co., St. Louis, MO; nonessential amino acids (0.1 mM), L-glutamine (2.0 mM), streptomycin (100 pg/ml), and penicillin (100 units/ml), which were purchased from GIBCO; and dextrose (2.7 mg/ml) obtained from Fisher Scientific. This medium was developed for human uroepithelial cells (Reznikoff et al., 1983, 1987). For subculturing, gelatin-coated T-75 flasks were used for primary cultures. After 10 days of primary culture, cells were dispersed for serial passage with 0.01% EDTA (Sigma Chemical Co.) and 0.25% trypsin (GIBCO) dissolved in HBSS (GIBCO) and added to gelatin-coated 35 mm plates as described above. Experiments were conducted with 80-90% confluent cultures (3-5 x lo5 cells per 35 mm plate). Growth media were aspirated and cells washed twice with 1ml of HBSS. Serum-free alpha-MEM was then added with or without test agents as indicated in the Results. Cells were incubated at 37°C in 5% C02-95%air. Test agents were arachidonic acid purchased from Nu-Chek Prep, Inc., Elysian, MN; calcium ionophore A23187 and staurosporine from CalBiochem, La Jolla, CA; and cycloheximide, aspirin, and TPA from Sigma Chemical Co., St. Louis, MO. Each agonist was used at a maximally effective concentration previously determined (Wong et al., 1989; Zenser e t al., 1990).
Radioimmunoassay of prostaglandin E, The PGE, content of the media was measured by double-antibody radioimmunoassay (Zenser and Davis, 1978). Rabbit antiserum to PGE, was obtained from Regis Chemical Co., Morton Grove, IL. Tritium-labeled PGE, was purchased from Du Pont NEN Research Products, Boston, MA; and goat antiserum to rabbit gamma globulins was purchased from Antibodies Inc., Davis, CA. Each condition was evaluated in two or three separate experiments. In each experiment, conditions were evaluated in a t least triplicate plates. The medium from each plate was analyzed in duplicate, and the value of duplicate determinations averaged and considered as a n N of one. Data were expressed as mean SEM of ng PGE, per lo5 cells. Differences between data were evaluated statistically by the Student's t-test for unpaired observations.
*
Prostaglandin H synthase assay Following a 24-hr incubation of cells in the absence and presence of TPA, media were saved for PGE, radioimmunoassay and microsomes were prepared for direct assessment of synthase activity (Brown et al., 1980). All subsequent procedures were performed a t 4°C. Cells were scraped into tubes, centrifuged at 2,000 rpm for 10 min to pellet cells, and were resuspended in buffer and
ZENSER ET AL.
216 50
40 h
m
T
0B A S A L ARACHIDONIC
PGE, PIIS ACTIVITY
E
2
1200
\
400
2
0
e 900
*" 30 0
?
k
r(
\
300
v
N 3
20
w 0 a
W
a >
v
