0021-972X/79/4905-0737$02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright© 1979 by The Endocrine Society
Vol. 49, No. 5 Printed in U.S.A.
In Vitro Effect of Dopamine and L-Dopa on Prolactin and Growth Hormone Release from Human Pituitary Adenomas* F. PEILLON, F. CESSELIN, D. BRESSION, N. ZYGELMAN, A. M. BRANDI, A. NOUSBAUM, AND A. MAUBORGNE INSERM FRA 33, Physio-Pathologie de VHypophyse (F.P., D.B., A.M.B., A.N.), et Service de Biochimie Medicate (F.C., N.Z., AM.), Faculte de Me'decine Pitie'-Salpetriere, 75634 Paris cedex 13, France
ABSTRACT. To determine the site of action of dopaminergic drugs on human PRL and GH release from pituitary adenomas, five PRL- and five GH-secreting adenomas were incubated with and without dopamine and L-dopa. Bromocriptine was also tested in order to compare its effect to that of the other drugs. In all of the experiments except one, a decrease of PRL, which was often statistically significant, was observed. When pooling the results of the PRL-secreting adenomas, the mean levels of PRL with dopamine, L-dopa, and bromocriptine were, respectively, 49%, 55%, and 60% of the control levels. In the GHsecreting adenomas, they were 60%, 67%, and 55% of that of the control. For GH, a decrease of the release was observed in four out of five GH-secreting adenomas. When pooling the results from these tumors, the mean levels of GH with dopamine, L-dopa, and bromocriptine were, respectively, 63%, 76%, and 64% of the
control levels. In one case, a significant increase of GH was observed with the three dopaminergic drugs. This study produced the following conclusions. 1) Dopamine acts directly on PRL and GH release from human pituitary adenomas; in vitro, L-dopa effects are similar (its action probably occurs after conversion to catecholamines). These observations strongly suggest the presence of dopaminergic membrane receptors on human lactotroph and somatotroph adenomatous pituitary cells. 2) In vitro hormonal results are in good agreement with in vivo dynamic tests using L-dopa and bromocriptine. 3) The paradoxical effect of dopaminergic drugs on GH secretion in acromegalic patients may be attributed to modified dopamine membrane receptors. However, the paradoxical response is not a constant feature in acromegaly, and its mechanism needs further investigations. (J Clin Endocrinol Metab 49: 737,1979)
I
T IS well known that dopaminergic agents influence PRL and GH secretion in man. In normal subjects, the administration of L-dopa is followed by an increase in GH release and a decrease in PRL secretion (1, 2). Dopamine also exerts an inhibitory effect on PRL secretion, but its effect on GH secretion is still controversial (3-5). In hyperprolactinemic subjects, the administration of L-dopa and dopamine is followed by a decrease in PRL release (2, 3), while in acromegaly, L-dopa and dopamine may exert an inhibitory effect on GH secretion, in contrast to normal men (6). Whether these drugs act at the hypothalamic and/or pituitary level is still unclear. Studies in rats have shown that the effect of catecholamines on the release of PRL was mediated through the release of an hypothalamic inhibiting factor (7, 8). On the other hand, dopamine and L-dopa have a direct inhibitory effect on PRL secretion from the rat pituitary in vitro (9-11) and in the stalk-sectioned rhesus monkey (12).
The observation that dopamine and L-dopa suppress TRH-induced PRL secretion in man (13, 14) is also in favor of a direct effect at the pituitary level. Dopamine and bromocriptine were unable to affect GH release from normal rat pituitaries in vitro (11), although ergot derivatives directly inhibited GH secretion by rat pituitary tumor tissue (15). To further delineate the site of action of dopamine and L-dopa in man, we studied their effect on PRL and GH secretion from human pituitary adenomas. Recently, two reports demonstrated the inhibitory effect of bromocriptine on PRL and GH release by GH-secreting adenomas in vitro (16, 17). Bromocriptine was also used in our study to evaluate whether there was a similar effect of the three dopaminergic drugs on PRL and/or GH release from the same adenomas. Materials and Methods Ten subjects (five with PRL-secreting adenomas and five with acromegaly) were tested with dopaminergic drugs before surgery. A test with L-dopa (0.5 g given orally) with plasma samples at -15,0, 60,90,120, and 180 min was carried out in all patients. In five cases (two PRL- and three GH-secreting adenomas), 2.5 mg of bromocriptine were administered orally and
Received November 28,1978. Address requests for reprints to F. Peillon, M.D., INSERM FRA 33, CHU Pitie-Salpetriere, 105 Bd de l'Hopital, 75634 Paris cedex 13, France. * This work was supported by grants from INSERM (CRAT 49 77 81) and from CNRS (ERA 484). 737
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 05 January 2016. at 07:31 For personal use only. No other uses without permission. . All rights reserved.
PEILLON ET AL.
738
blood specimens were collected at 30-min intervals for 3 h. Serum PRL and GH responses were defined as positive when a decrease of at least 50% below the mean of the control value was achieved. Culture method After surgery, most of the tumor tissue was placed into culture medium (medium 199-buffered Hepes), as described (18), but no fetal calf serum was added. Fragments from each tumor were pooled and divided into pieces of approximately 1 mm3 in size. Five tumor pieces were mounted with cataract knives onto a strip of moistened lens paper placed on a grid of
JCE&M • 1979 Vol 49 • No 5
stainless steel. The grid supporting the explants was placed in a culture chamber of borosilicate glass 30 mm in diameter filled with culture medium to the level of the grids (usually 1.5 ml). The incubation was performed at 37 C, and all of the experiments were made in triplicate. After 30 min of preincubation, the medium was discarded and replaced with 1.5 ml of fresh medium with or without the following dopaminergic drugs: dopamine (5.2 X 10"5, 5.2 X 10"6, 5.2 x 10"7 M), L-dopa (10~4, 10"5, and 10~6 M) and bromocriptine (6.5 X 10~4, 6.5 X 10"5, 6.5 X 10~6 M). According to the size of the tumor, the experiments were run with all of the drugs and concentrations or only one or two of them (see Tables 1 and 2). Each assay point was done in triplicate.
TABLE 1. Effect of dopamine (D), L-dopa (L), and bromocriptine (CB) on in vitro PRL release from five prolactinomas and five GH-secreting adenomas (I) and nadir plasma PRL and GH levels [percentage of the control values] during L-dopa and bromocriptine tests performed in the patients before surgery (II)
Prolactinomas In vitro (I) C D L CB In vivo (II) L CB
GH-secreting adenomas In vitro (I) C D L CB In vivo (II) L
CB
1 (ng/jug protein)
2 (ng/fig protein)
3 (ng/mg tissue)
426 ±54
92 ±24 56 ±7° 51 ± ll c 56 ± 10°
714 ± 311 265 ± 141" 445 ±264 428 ± 376
38 23
47 39
288 ± 37*
20
7 (pg/jug protein)
8 (pg//ig protein)
39 ±17 19 ±7° 31 ± 8 16 ± 6"
34 ±24 12±3 C 19 ±11
10 ± 3 10 ±18 6±l c 5±l c
52
3227 ± 925 1571 ± 640°
5 (pg/jug protein)
596 ± 220 286 ± 123 254 ± 94C
20
39
6 (ng/jug protein)
19 21
4 (ng/mg tissue)
9
(pg/^g protein)
10 (pg/mg tissue)
56 ±35
764 ± 191 630 ± 220 622 ± 198 771 ± 543
20 ± 16C 33 ±36 17 ± 14e
37
36 45
80
87
PRL and GH values given are the mean ± SD. P< 0.010 vs. control (C). * P < 0.050 vs. control (C). C P < 0.025 vs. control (C).
a
TABLE 2. Effect of dopamine (D), L-dopa (L), and bromocriptine (CB) on in vitro GH release from five GH-secreting adenomas (I) and nadir plasma GH levels [percentage of the control values] during L-dopa and bromocriptine tests performed in the patients before surgery (II) Tumor no.
In vitro (I) C D L
CB
6 (ng/jug protein)
7 (ng/ftg protein)
8 (ng/jug protein)
9 (ng/fig protein)
10 (ng/mg tissue)
100 ±32 53 ± 18° 55 ± 11" 64 ±35
22 ±30 14 ± 8 17 ±11
191 ± 141 118 ± 80 176 ±38 119 ± 75
36 ±12 54 ±14* 70 ± 15" 58 ±6°
90 ±30 65 ± 11* 72 ±11 60 ± 19*
50 33
143 115
100 42
In vivo (II)
L CB
23 31
42
PRL values given are the mean ± SD. ° P < 0.01 vs. control (C). "P < 0.05 vs. control (C).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 05 January 2016. at 07:31 For personal use only. No other uses without permission. . All rights reserved.
DOPAMINERGIC DRUGS, HUMAN PRL, AND GH IN VITRO After 4 h of incubation, media were removed and frozen at -20 C until assayed for PRL and GH. RIA of PRL and GH were carried out by a double antibody technique, as previously described (18). The assays were made at several dilutions, each in duplicate, in order to check the parallelism between the competition curves so obtained and the standard curves. The results reported here are the mean of these determinations. The coefficients of variation within and between assays, were 7% and 9% for PRL and 5% and 12% for GH, respectively. The dopaminergic drugs were kindly provided by Sigma Chemical Co. (dopamine; St. Louis, MO), Roche Scientific Corp. (L-dopa; France), and Sandoz (bromocriptine; Switzerland). For electron microscopic studies, the explants of two PRLand one GH-secreting adenomas were weighed at the end of the incubation, while in the other cases, tissue proteins were determined according to the method of Lowry (19). Experimental data were analyzed statistically by Student's t test and variance analysis.
Results Tables 1 and 2 display the in vitro as well as the in vivo data. As no statistically significant difference was observed in vitro between the means of the results obtained with the different concentrations of each dopaminergic agent (variance analysis), pooled data for the same drug were compared to control values (i.e. experiments without dopaminergic drugs). Results on PRL secretion (Table 1) In all of the experiments except one (tumor 10), a decrease in PRL was observed when the dopaminergic drugs were added. Moreover, despite the dispersion of the data obtained by replicate incubations of the tumor fragments, the inhibitory effect was often statistically significant. When pooling the results from all of the tumors (except tumor 10), the mean level of PRL was GH-SECRETING ADENOMAS
PRL-SECRETING ADENOMAS
ioor m
5 CASES
75
75
50
50
25
ffl
4 CASES
100
T
f
r
25 •'
Vol. 49, No. 5 Printed in U.S.A.
In Vitro Effect of Dopamine and L-Dopa on Prolactin and Growth Hormone Release from Human Pituitary Adenomas* F. PEILLON, F. CESSELIN, D. BRESSION, N. ZYGELMAN, A. M. BRANDI, A. NOUSBAUM, AND A. MAUBORGNE INSERM FRA 33, Physio-Pathologie de VHypophyse (F.P., D.B., A.M.B., A.N.), et Service de Biochimie Medicate (F.C., N.Z., AM.), Faculte de Me'decine Pitie'-Salpetriere, 75634 Paris cedex 13, France
ABSTRACT. To determine the site of action of dopaminergic drugs on human PRL and GH release from pituitary adenomas, five PRL- and five GH-secreting adenomas were incubated with and without dopamine and L-dopa. Bromocriptine was also tested in order to compare its effect to that of the other drugs. In all of the experiments except one, a decrease of PRL, which was often statistically significant, was observed. When pooling the results of the PRL-secreting adenomas, the mean levels of PRL with dopamine, L-dopa, and bromocriptine were, respectively, 49%, 55%, and 60% of the control levels. In the GHsecreting adenomas, they were 60%, 67%, and 55% of that of the control. For GH, a decrease of the release was observed in four out of five GH-secreting adenomas. When pooling the results from these tumors, the mean levels of GH with dopamine, L-dopa, and bromocriptine were, respectively, 63%, 76%, and 64% of the
control levels. In one case, a significant increase of GH was observed with the three dopaminergic drugs. This study produced the following conclusions. 1) Dopamine acts directly on PRL and GH release from human pituitary adenomas; in vitro, L-dopa effects are similar (its action probably occurs after conversion to catecholamines). These observations strongly suggest the presence of dopaminergic membrane receptors on human lactotroph and somatotroph adenomatous pituitary cells. 2) In vitro hormonal results are in good agreement with in vivo dynamic tests using L-dopa and bromocriptine. 3) The paradoxical effect of dopaminergic drugs on GH secretion in acromegalic patients may be attributed to modified dopamine membrane receptors. However, the paradoxical response is not a constant feature in acromegaly, and its mechanism needs further investigations. (J Clin Endocrinol Metab 49: 737,1979)
I
T IS well known that dopaminergic agents influence PRL and GH secretion in man. In normal subjects, the administration of L-dopa is followed by an increase in GH release and a decrease in PRL secretion (1, 2). Dopamine also exerts an inhibitory effect on PRL secretion, but its effect on GH secretion is still controversial (3-5). In hyperprolactinemic subjects, the administration of L-dopa and dopamine is followed by a decrease in PRL release (2, 3), while in acromegaly, L-dopa and dopamine may exert an inhibitory effect on GH secretion, in contrast to normal men (6). Whether these drugs act at the hypothalamic and/or pituitary level is still unclear. Studies in rats have shown that the effect of catecholamines on the release of PRL was mediated through the release of an hypothalamic inhibiting factor (7, 8). On the other hand, dopamine and L-dopa have a direct inhibitory effect on PRL secretion from the rat pituitary in vitro (9-11) and in the stalk-sectioned rhesus monkey (12).
The observation that dopamine and L-dopa suppress TRH-induced PRL secretion in man (13, 14) is also in favor of a direct effect at the pituitary level. Dopamine and bromocriptine were unable to affect GH release from normal rat pituitaries in vitro (11), although ergot derivatives directly inhibited GH secretion by rat pituitary tumor tissue (15). To further delineate the site of action of dopamine and L-dopa in man, we studied their effect on PRL and GH secretion from human pituitary adenomas. Recently, two reports demonstrated the inhibitory effect of bromocriptine on PRL and GH release by GH-secreting adenomas in vitro (16, 17). Bromocriptine was also used in our study to evaluate whether there was a similar effect of the three dopaminergic drugs on PRL and/or GH release from the same adenomas. Materials and Methods Ten subjects (five with PRL-secreting adenomas and five with acromegaly) were tested with dopaminergic drugs before surgery. A test with L-dopa (0.5 g given orally) with plasma samples at -15,0, 60,90,120, and 180 min was carried out in all patients. In five cases (two PRL- and three GH-secreting adenomas), 2.5 mg of bromocriptine were administered orally and
Received November 28,1978. Address requests for reprints to F. Peillon, M.D., INSERM FRA 33, CHU Pitie-Salpetriere, 105 Bd de l'Hopital, 75634 Paris cedex 13, France. * This work was supported by grants from INSERM (CRAT 49 77 81) and from CNRS (ERA 484). 737
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 05 January 2016. at 07:31 For personal use only. No other uses without permission. . All rights reserved.
PEILLON ET AL.
738
blood specimens were collected at 30-min intervals for 3 h. Serum PRL and GH responses were defined as positive when a decrease of at least 50% below the mean of the control value was achieved. Culture method After surgery, most of the tumor tissue was placed into culture medium (medium 199-buffered Hepes), as described (18), but no fetal calf serum was added. Fragments from each tumor were pooled and divided into pieces of approximately 1 mm3 in size. Five tumor pieces were mounted with cataract knives onto a strip of moistened lens paper placed on a grid of
JCE&M • 1979 Vol 49 • No 5
stainless steel. The grid supporting the explants was placed in a culture chamber of borosilicate glass 30 mm in diameter filled with culture medium to the level of the grids (usually 1.5 ml). The incubation was performed at 37 C, and all of the experiments were made in triplicate. After 30 min of preincubation, the medium was discarded and replaced with 1.5 ml of fresh medium with or without the following dopaminergic drugs: dopamine (5.2 X 10"5, 5.2 X 10"6, 5.2 x 10"7 M), L-dopa (10~4, 10"5, and 10~6 M) and bromocriptine (6.5 X 10~4, 6.5 X 10"5, 6.5 X 10~6 M). According to the size of the tumor, the experiments were run with all of the drugs and concentrations or only one or two of them (see Tables 1 and 2). Each assay point was done in triplicate.
TABLE 1. Effect of dopamine (D), L-dopa (L), and bromocriptine (CB) on in vitro PRL release from five prolactinomas and five GH-secreting adenomas (I) and nadir plasma PRL and GH levels [percentage of the control values] during L-dopa and bromocriptine tests performed in the patients before surgery (II)
Prolactinomas In vitro (I) C D L CB In vivo (II) L CB
GH-secreting adenomas In vitro (I) C D L CB In vivo (II) L
CB
1 (ng/jug protein)
2 (ng/fig protein)
3 (ng/mg tissue)
426 ±54
92 ±24 56 ±7° 51 ± ll c 56 ± 10°
714 ± 311 265 ± 141" 445 ±264 428 ± 376
38 23
47 39
288 ± 37*
20
7 (pg/jug protein)
8 (pg//ig protein)
39 ±17 19 ±7° 31 ± 8 16 ± 6"
34 ±24 12±3 C 19 ±11
10 ± 3 10 ±18 6±l c 5±l c
52
3227 ± 925 1571 ± 640°
5 (pg/jug protein)
596 ± 220 286 ± 123 254 ± 94C
20
39
6 (ng/jug protein)
19 21
4 (ng/mg tissue)
9
(pg/^g protein)
10 (pg/mg tissue)
56 ±35
764 ± 191 630 ± 220 622 ± 198 771 ± 543
20 ± 16C 33 ±36 17 ± 14e
37
36 45
80
87
PRL and GH values given are the mean ± SD. P< 0.010 vs. control (C). * P < 0.050 vs. control (C). C P < 0.025 vs. control (C).
a
TABLE 2. Effect of dopamine (D), L-dopa (L), and bromocriptine (CB) on in vitro GH release from five GH-secreting adenomas (I) and nadir plasma GH levels [percentage of the control values] during L-dopa and bromocriptine tests performed in the patients before surgery (II) Tumor no.
In vitro (I) C D L
CB
6 (ng/jug protein)
7 (ng/ftg protein)
8 (ng/jug protein)
9 (ng/fig protein)
10 (ng/mg tissue)
100 ±32 53 ± 18° 55 ± 11" 64 ±35
22 ±30 14 ± 8 17 ±11
191 ± 141 118 ± 80 176 ±38 119 ± 75
36 ±12 54 ±14* 70 ± 15" 58 ±6°
90 ±30 65 ± 11* 72 ±11 60 ± 19*
50 33
143 115
100 42
In vivo (II)
L CB
23 31
42
PRL values given are the mean ± SD. ° P < 0.01 vs. control (C). "P < 0.05 vs. control (C).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 05 January 2016. at 07:31 For personal use only. No other uses without permission. . All rights reserved.
DOPAMINERGIC DRUGS, HUMAN PRL, AND GH IN VITRO After 4 h of incubation, media were removed and frozen at -20 C until assayed for PRL and GH. RIA of PRL and GH were carried out by a double antibody technique, as previously described (18). The assays were made at several dilutions, each in duplicate, in order to check the parallelism between the competition curves so obtained and the standard curves. The results reported here are the mean of these determinations. The coefficients of variation within and between assays, were 7% and 9% for PRL and 5% and 12% for GH, respectively. The dopaminergic drugs were kindly provided by Sigma Chemical Co. (dopamine; St. Louis, MO), Roche Scientific Corp. (L-dopa; France), and Sandoz (bromocriptine; Switzerland). For electron microscopic studies, the explants of two PRLand one GH-secreting adenomas were weighed at the end of the incubation, while in the other cases, tissue proteins were determined according to the method of Lowry (19). Experimental data were analyzed statistically by Student's t test and variance analysis.
Results Tables 1 and 2 display the in vitro as well as the in vivo data. As no statistically significant difference was observed in vitro between the means of the results obtained with the different concentrations of each dopaminergic agent (variance analysis), pooled data for the same drug were compared to control values (i.e. experiments without dopaminergic drugs). Results on PRL secretion (Table 1) In all of the experiments except one (tumor 10), a decrease in PRL was observed when the dopaminergic drugs were added. Moreover, despite the dispersion of the data obtained by replicate incubations of the tumor fragments, the inhibitory effect was often statistically significant. When pooling the results from all of the tumors (except tumor 10), the mean level of PRL was GH-SECRETING ADENOMAS
PRL-SECRETING ADENOMAS
ioor m
5 CASES
75
75
50
50
25
ffl
4 CASES
100
T
f
r
25 •'
