0013-7227/90/1274-1811$02.00/0 Endocrinology Copyright © 1990 by The Endocrine Society
Vol. 127, No. 4 Printed in U.S.A.
Biphasic Action of Forskolin on Growth Hormone and Prolactin Secretion by Rat Anterior Pituitary Cells in Vitro* MARTA SZABO, NEIL E. STAIB, BARBARA J. COLLINS, AND LEONA CUTTLER Department of Medicine, Michael Reese Hospital and Medical Center, University of Illinois College of Medicine (M.S., N.E.S.), and the Department of Pediatrics, Wyler Children's Hospital, University of Chicago Pritzker School of Medicine (B.J.C., L.C.), Chicago, Illinois 60616; and the Department of Pediatrics, Case Western Reserve University (L.C.), Cleveland, Ohio 44106
ABSTRACT. To assess the role of cAMP-mediated signal transduction processes in mediation of secretagogue-stimulated GH release, we examined the dose-related effects of the diterpene adenylate cyclase activator forskolin (FSK) in primary monolayer cultures of rat adenohypophyseal cells. In cell cultures prepared from both immature (12 days old) and adult (6 weeks to 4 months old) male or female rats, the dose-related stimulation of GH release by FSK was biphasic. With increasing FSK concentrations from 0.03-3.16 nM, GH release increased progressively to maximal values of 442 ± 19% and 303 ± 10% of basal release in cells from immature and adult rats, respectively. FSK concentrations above 3.16 /uM induced progressively diminished GH responses, with net inhibition to below basal release evident at 100 ^M FSK. FSK stimulated PRL release to a lesser degree than it did GH release; the PRL response to FSK was also biphasic. When maximal stimulatory concentrations (Ema J of FSK and GH-releasing factor (GRF; 10 nM) were added in combination, the GH response was significantly less than the individual response to either secretagogue alone. In response to FSK alone, GRF alone, and FSK plus GRF, GH release was 478 ± 7%, 583 ± 11%, and 244 ± 5%; 278 ± 4%, 283 ± 3%, and 175 ± 2%; and 299 ± 12%, 351 ± 5%, and 191 ± 17% of basal release in cells from 12-day-old, adult male, and adult female rats, respectively (P < 0.01 for all responses to combined addition vs. the individual responses). Submaximal stimulatory concentrations of GRF added in combination with submaximal FSK elicited partially additive GH responses; the GH response to Emax GRF, on the other hand, was inhibited in a dose-related manner by all con-
S
INCE the first demonstration by Seamon and coworkers (1) that the plant diterpene forskolin (FSK) is a direct activator of adenylate cyclase in intact
Received April 20, 1990. Address all correspondence and requests for reprints to: Marta Szabo, Ph.D., Division of Endocrinology and Metabolism, Department of Medicine, Michael Reese Hospital and Medical Center, Lake Shore Drive at 31st Street, Chicago, Illinois 60616. * This work was supported in part by NIH Grant P-60-DK-20595 and a grant from the Michael Reese Medical Research Institute Council (to M.S.) and by NIH Grant DK-40221 and the Basil O'Connor Research Starter Award (no. 5-609) from the March of Dimes Birth Defects Foundation (to L.C.). Presented in part at the 70th Annual Meeting of The Endocrine Society, June 1988, New Orleans, LA (Abstract 960).
centrations of FSK that by themselves were stimulatory. The GH responses were also suppressed when Emox FSK was added to cultured cells of 12-day-old rats in combination with Ema* cholera toxin (2.5 ng/ml) or prostaglandin E2 (10 /uM), agents whose actions, like that of GRF, involve adenylate cyclase activation. In contrast, FSK did not suppress but in most cases augmented the maximal GH responses to secretagogues whose action is independent of adenylate cyclase activation: (Bu)2cAMP (0.5 mM), TRH (100 nM), phorbol myristate acetate (50 nM), the Ca2+ ionophore A23187 (250 MM), and the dihydropyridine Ca2+ channel agonist BAY K8644 (10 nM). Indeed, combined addition of FSK with the latter two agents resulted in synergistic stimulation. Pretreatment of the cultured cells with pertussis toxin (1-100 ng/ml) for 24 h attenuated the suppressive effect of FSK on GRF-induced GH release, implicating participation of a Gi protein in this phenomenon. Participation of a Gi protein was also suggested by the finding that in cell cultures from adult male rats, GH release induced by combined GRF and FSK was significantly less susceptible to suppression by somatostatin than either GRF- or FSK-induced GH release. The suppressed GH responses obtained upon combined addition of GRF and FSK could not, however, be explained by direct inhibition of GRF-stimulated adenylate cyclase by FSK, as the intracellular cAMP responses were additive at all secretagogue concentrations tested; this finding suggests that FSK exerts its inhibitory action at a site distal to cAMP generation. {Endocrinology 127: 18111817,1990)
cells, FSK has been widely used to probe the role of cAMP in mediation of cellular responses to extracellular signals. Early recognition that GH-releasing factor (GRF) stimulates the release of GH through a cAMPdependent action was based in part on the demonstration by Brazeau et al. (2) that FSK, which by itself elicited a dose-dependent stimulation of GH release in rat anterior pituitary cell cultures, did not augment the response to a maximal stimulatory concentration (Emax) of GRF. We adopted a similar argument in studying the interaction of FSK with TRH to obtain additional evidence for a cAMP-independent action of TRH in the somatotroph, after having demonstrated additive and synergistic in-
1811
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 06 November 2014. at 14:31 For personal use only. No other uses without permission. . All rights reserved.
1812
BIPHASIC ACTION OF FSK ON GH AND PRL RELEASE
teraction of Emax TRH with GRF and (Bu)2cAMP in anterior pituitary cells of hypothyroid adult rats (3) and 12-day-old rats (4). An initial experiment, aimed at comparing the GH response to combined TRH and FSK with the expected nonadditive response to GRF and FSK, however, revealed an unexpected inhibition of the maximal GH responses to GRF and FSK when the two agents were added concurrently to cultured cells of 12-day-old rats. This observation, which was subsequently confirmed in numerous experiments with anterior pituitary cells of both neonatal and adult rats, prompted us to reinvestigate the dose-related stimulation of GH release by FSK and the interaction of FSK with various concentrations of GRF and with several other GH secretagogues whose actions are either cAMP mediated or independent of cAMP generation. We describe here a previously unrecognized biphasic action of FSK on basal and GRFinduced GH release as well as on basal PRL release.
Endo • 1990 Vol 127 • No 4
cultures from 12-day-old and adult rat pituitaries, respectively. The yield of viable cells per gland ranged from 0.8-1.2 X 106 for 12-day-old rats to 2-5 X 106 for adult rats. FSK was diluted to the indicated concentrations from a 10-mM solution, and A23187 from a 25-mM solution in dimethylsulfoxide; control incubations with the highest resulting dimethylsulfoxide concentration (1%) revealed no effect of the solvent by itself on hormone secretion. PMA and PGE2 were diluted from 16- and 1-mM stocks in ethanol; the resulting maximal ethanol concentration (1%) was also without effect on basal and stimulated hormone release. All other agents were diluted from aqueous buffered stocks. Pretreatment with PT was begun on the third day of culture, 24 h before incubation with the secretagogues. cAMP measurement was performed in cells from adult male rats after 2 days of primary culture to minimize fibroblast growth and cAMP accumulation in cells other than somatotrophs. In this experiment, incubations were terminated at 60 min, when intracellular cAMP levels are near their peak (6, 7). RIAs
Materials and Methods Materials Rat GRF, somatostatin (SRIF), and TRH were obtained from Peninsula Laboratories (Belmont, CA); FSK and cholera toxin (CT) were purchased from Calbiochem-Behring (La Jolla, CA). BAY K8644 was kindly provided by Dr. Alexander Scriabine, Miles, Inc. (West Haven, CT). Pertussis toxin (PT), prostaglandin E2 (PGE2), (Bu)2cAMP, phorbol myristate acetate (PMA), and A23187 were obtained from Sigma Chemical Co. (St. Louis, MO). Other reagents, culture media, and serum supplements were obtained from routine commercial sources, as described previously (5). Animals Adult male Sprague-Dawley rats, timed pregnant female rats, and female rats with their litters of a defined birth date were purchased from Charles River Laboratories, Inc. (Wilmington, MA). They were maintained in temperature-controlled quarters, with lights on between 0600-1800 h, on standard rat chow and tap water ad libitum, for at least 1 week before the experiment. Newborn rats of both sexes were kept with their mothers until they were used on postnatal day 12. Rats were killed by decapitation; the skull was then opened for rapid excision of the anterior pituitary, which was kept in chilled Dulbecco's Modified Eagle's Medium with 25 mM HEPES and 0.3% BSA until enzymatic dissociation. Anterior pituitary cell culture and incubation Anterior pituitary tissue was dissociated with trypsin-EDTA, and cells were maintained in primary monolayer culture for 4 days and incubated with the indicated concentrations of secretagogues for 3 h, as detailed previously (5). Briefly, anterior pituitary cell suspensions were distributed in multiwell culture plates with 16-mm diameter wells (Corning Glass Works, Corning, NY) at densities of 1-2 x 105 and 0.5-1 x 105 cells/well for
Rat GH and PRL concentrations in the incubation medium were determined by double antibody RIAs, as previously described (8). Hormone concentrations are expressed in terms of the NIDDK rat GH and rat PRL RP-1 standards. Intracellular cAMP was extracted from cell monolayers with 6% trichloroacetic acid and assayed according to the method of BerryKravis and Dawson (9). Statistical analyses All values are expressed as the mean ± SEM. To control for differences in basal release among independent experiments, GH release or suppression is expressed as a percentage of basal or stimulated values. P values for significance of differences among groups were calculated by Tukey's multiple comparison procedure with the SYSTAT statistical software package (10). A P value of 0.05 was considered significant. EC50 values were calculated by the four-parameter logistic equation (11).
Results Dose-related effects of FSK alone and in combination with GRF In initial studies of the interaction between FSK and GRF on GH release, 10 pM FSK was added with Emax GRF (10 nM) to a 4-day culture of adenohypophyseal cells from 12-day-old rats (Table 1). In these preliminary experiments, the concentration of FSK selected for maximal stimulation was based on previously reported data (2,12). Unexpectedly, we found the magnitude of the GH response to combined GRF and FSK to be significantly less (P < 0.001) than that to either secretagogue alone, a finding consistently confirmed in subsequent experiments; similar suppression of a lesser magnitude was also demonstrated in cells from adult animals of both sexes. These findings prompted us to reevaluate the dose-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 06 November 2014. at 14:31 For personal use only. No other uses without permission. . All rights reserved.
BIPHASIC ACTION OF FSK ON GH AND PRL RELEASE TABLE 1. Effects of maximal stimulatory concentrations of GRF and FSK alone and in combination on GH release in rat adenohypophyseal cell cultures rGH release (ng/ml-3 h)° Addition
GRF (10 nM) FSK (10 /tM) GRF + FSK
12-day-old
Adult male
299 ± 1744 ± 1430 ± 729 ±
720 ± 2039 ± 2002 ± 1263 ±
P for GRF + FSK us. GRF alone FSK alone
10 32 20 15
Vol. 127, No. 4 Printed in U.S.A.
Biphasic Action of Forskolin on Growth Hormone and Prolactin Secretion by Rat Anterior Pituitary Cells in Vitro* MARTA SZABO, NEIL E. STAIB, BARBARA J. COLLINS, AND LEONA CUTTLER Department of Medicine, Michael Reese Hospital and Medical Center, University of Illinois College of Medicine (M.S., N.E.S.), and the Department of Pediatrics, Wyler Children's Hospital, University of Chicago Pritzker School of Medicine (B.J.C., L.C.), Chicago, Illinois 60616; and the Department of Pediatrics, Case Western Reserve University (L.C.), Cleveland, Ohio 44106
ABSTRACT. To assess the role of cAMP-mediated signal transduction processes in mediation of secretagogue-stimulated GH release, we examined the dose-related effects of the diterpene adenylate cyclase activator forskolin (FSK) in primary monolayer cultures of rat adenohypophyseal cells. In cell cultures prepared from both immature (12 days old) and adult (6 weeks to 4 months old) male or female rats, the dose-related stimulation of GH release by FSK was biphasic. With increasing FSK concentrations from 0.03-3.16 nM, GH release increased progressively to maximal values of 442 ± 19% and 303 ± 10% of basal release in cells from immature and adult rats, respectively. FSK concentrations above 3.16 /uM induced progressively diminished GH responses, with net inhibition to below basal release evident at 100 ^M FSK. FSK stimulated PRL release to a lesser degree than it did GH release; the PRL response to FSK was also biphasic. When maximal stimulatory concentrations (Ema J of FSK and GH-releasing factor (GRF; 10 nM) were added in combination, the GH response was significantly less than the individual response to either secretagogue alone. In response to FSK alone, GRF alone, and FSK plus GRF, GH release was 478 ± 7%, 583 ± 11%, and 244 ± 5%; 278 ± 4%, 283 ± 3%, and 175 ± 2%; and 299 ± 12%, 351 ± 5%, and 191 ± 17% of basal release in cells from 12-day-old, adult male, and adult female rats, respectively (P < 0.01 for all responses to combined addition vs. the individual responses). Submaximal stimulatory concentrations of GRF added in combination with submaximal FSK elicited partially additive GH responses; the GH response to Emax GRF, on the other hand, was inhibited in a dose-related manner by all con-
S
INCE the first demonstration by Seamon and coworkers (1) that the plant diterpene forskolin (FSK) is a direct activator of adenylate cyclase in intact
Received April 20, 1990. Address all correspondence and requests for reprints to: Marta Szabo, Ph.D., Division of Endocrinology and Metabolism, Department of Medicine, Michael Reese Hospital and Medical Center, Lake Shore Drive at 31st Street, Chicago, Illinois 60616. * This work was supported in part by NIH Grant P-60-DK-20595 and a grant from the Michael Reese Medical Research Institute Council (to M.S.) and by NIH Grant DK-40221 and the Basil O'Connor Research Starter Award (no. 5-609) from the March of Dimes Birth Defects Foundation (to L.C.). Presented in part at the 70th Annual Meeting of The Endocrine Society, June 1988, New Orleans, LA (Abstract 960).
centrations of FSK that by themselves were stimulatory. The GH responses were also suppressed when Emox FSK was added to cultured cells of 12-day-old rats in combination with Ema* cholera toxin (2.5 ng/ml) or prostaglandin E2 (10 /uM), agents whose actions, like that of GRF, involve adenylate cyclase activation. In contrast, FSK did not suppress but in most cases augmented the maximal GH responses to secretagogues whose action is independent of adenylate cyclase activation: (Bu)2cAMP (0.5 mM), TRH (100 nM), phorbol myristate acetate (50 nM), the Ca2+ ionophore A23187 (250 MM), and the dihydropyridine Ca2+ channel agonist BAY K8644 (10 nM). Indeed, combined addition of FSK with the latter two agents resulted in synergistic stimulation. Pretreatment of the cultured cells with pertussis toxin (1-100 ng/ml) for 24 h attenuated the suppressive effect of FSK on GRF-induced GH release, implicating participation of a Gi protein in this phenomenon. Participation of a Gi protein was also suggested by the finding that in cell cultures from adult male rats, GH release induced by combined GRF and FSK was significantly less susceptible to suppression by somatostatin than either GRF- or FSK-induced GH release. The suppressed GH responses obtained upon combined addition of GRF and FSK could not, however, be explained by direct inhibition of GRF-stimulated adenylate cyclase by FSK, as the intracellular cAMP responses were additive at all secretagogue concentrations tested; this finding suggests that FSK exerts its inhibitory action at a site distal to cAMP generation. {Endocrinology 127: 18111817,1990)
cells, FSK has been widely used to probe the role of cAMP in mediation of cellular responses to extracellular signals. Early recognition that GH-releasing factor (GRF) stimulates the release of GH through a cAMPdependent action was based in part on the demonstration by Brazeau et al. (2) that FSK, which by itself elicited a dose-dependent stimulation of GH release in rat anterior pituitary cell cultures, did not augment the response to a maximal stimulatory concentration (Emax) of GRF. We adopted a similar argument in studying the interaction of FSK with TRH to obtain additional evidence for a cAMP-independent action of TRH in the somatotroph, after having demonstrated additive and synergistic in-
1811
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 06 November 2014. at 14:31 For personal use only. No other uses without permission. . All rights reserved.
1812
BIPHASIC ACTION OF FSK ON GH AND PRL RELEASE
teraction of Emax TRH with GRF and (Bu)2cAMP in anterior pituitary cells of hypothyroid adult rats (3) and 12-day-old rats (4). An initial experiment, aimed at comparing the GH response to combined TRH and FSK with the expected nonadditive response to GRF and FSK, however, revealed an unexpected inhibition of the maximal GH responses to GRF and FSK when the two agents were added concurrently to cultured cells of 12-day-old rats. This observation, which was subsequently confirmed in numerous experiments with anterior pituitary cells of both neonatal and adult rats, prompted us to reinvestigate the dose-related stimulation of GH release by FSK and the interaction of FSK with various concentrations of GRF and with several other GH secretagogues whose actions are either cAMP mediated or independent of cAMP generation. We describe here a previously unrecognized biphasic action of FSK on basal and GRFinduced GH release as well as on basal PRL release.
Endo • 1990 Vol 127 • No 4
cultures from 12-day-old and adult rat pituitaries, respectively. The yield of viable cells per gland ranged from 0.8-1.2 X 106 for 12-day-old rats to 2-5 X 106 for adult rats. FSK was diluted to the indicated concentrations from a 10-mM solution, and A23187 from a 25-mM solution in dimethylsulfoxide; control incubations with the highest resulting dimethylsulfoxide concentration (1%) revealed no effect of the solvent by itself on hormone secretion. PMA and PGE2 were diluted from 16- and 1-mM stocks in ethanol; the resulting maximal ethanol concentration (1%) was also without effect on basal and stimulated hormone release. All other agents were diluted from aqueous buffered stocks. Pretreatment with PT was begun on the third day of culture, 24 h before incubation with the secretagogues. cAMP measurement was performed in cells from adult male rats after 2 days of primary culture to minimize fibroblast growth and cAMP accumulation in cells other than somatotrophs. In this experiment, incubations were terminated at 60 min, when intracellular cAMP levels are near their peak (6, 7). RIAs
Materials and Methods Materials Rat GRF, somatostatin (SRIF), and TRH were obtained from Peninsula Laboratories (Belmont, CA); FSK and cholera toxin (CT) were purchased from Calbiochem-Behring (La Jolla, CA). BAY K8644 was kindly provided by Dr. Alexander Scriabine, Miles, Inc. (West Haven, CT). Pertussis toxin (PT), prostaglandin E2 (PGE2), (Bu)2cAMP, phorbol myristate acetate (PMA), and A23187 were obtained from Sigma Chemical Co. (St. Louis, MO). Other reagents, culture media, and serum supplements were obtained from routine commercial sources, as described previously (5). Animals Adult male Sprague-Dawley rats, timed pregnant female rats, and female rats with their litters of a defined birth date were purchased from Charles River Laboratories, Inc. (Wilmington, MA). They were maintained in temperature-controlled quarters, with lights on between 0600-1800 h, on standard rat chow and tap water ad libitum, for at least 1 week before the experiment. Newborn rats of both sexes were kept with their mothers until they were used on postnatal day 12. Rats were killed by decapitation; the skull was then opened for rapid excision of the anterior pituitary, which was kept in chilled Dulbecco's Modified Eagle's Medium with 25 mM HEPES and 0.3% BSA until enzymatic dissociation. Anterior pituitary cell culture and incubation Anterior pituitary tissue was dissociated with trypsin-EDTA, and cells were maintained in primary monolayer culture for 4 days and incubated with the indicated concentrations of secretagogues for 3 h, as detailed previously (5). Briefly, anterior pituitary cell suspensions were distributed in multiwell culture plates with 16-mm diameter wells (Corning Glass Works, Corning, NY) at densities of 1-2 x 105 and 0.5-1 x 105 cells/well for
Rat GH and PRL concentrations in the incubation medium were determined by double antibody RIAs, as previously described (8). Hormone concentrations are expressed in terms of the NIDDK rat GH and rat PRL RP-1 standards. Intracellular cAMP was extracted from cell monolayers with 6% trichloroacetic acid and assayed according to the method of BerryKravis and Dawson (9). Statistical analyses All values are expressed as the mean ± SEM. To control for differences in basal release among independent experiments, GH release or suppression is expressed as a percentage of basal or stimulated values. P values for significance of differences among groups were calculated by Tukey's multiple comparison procedure with the SYSTAT statistical software package (10). A P value of 0.05 was considered significant. EC50 values were calculated by the four-parameter logistic equation (11).
Results Dose-related effects of FSK alone and in combination with GRF In initial studies of the interaction between FSK and GRF on GH release, 10 pM FSK was added with Emax GRF (10 nM) to a 4-day culture of adenohypophyseal cells from 12-day-old rats (Table 1). In these preliminary experiments, the concentration of FSK selected for maximal stimulation was based on previously reported data (2,12). Unexpectedly, we found the magnitude of the GH response to combined GRF and FSK to be significantly less (P < 0.001) than that to either secretagogue alone, a finding consistently confirmed in subsequent experiments; similar suppression of a lesser magnitude was also demonstrated in cells from adult animals of both sexes. These findings prompted us to reevaluate the dose-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 06 November 2014. at 14:31 For personal use only. No other uses without permission. . All rights reserved.
BIPHASIC ACTION OF FSK ON GH AND PRL RELEASE TABLE 1. Effects of maximal stimulatory concentrations of GRF and FSK alone and in combination on GH release in rat adenohypophyseal cell cultures rGH release (ng/ml-3 h)° Addition
GRF (10 nM) FSK (10 /tM) GRF + FSK
12-day-old
Adult male
299 ± 1744 ± 1430 ± 729 ±
720 ± 2039 ± 2002 ± 1263 ±
P for GRF + FSK us. GRF alone FSK alone
10 32 20 15
