0021 -972X/78/4703-0480$02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1978 by The Endocrine Society
Vol. 47, No. 3 Printed in U.S.A.
Dopamine Affects Basal and Augmented Pituitary Hormone Secretion* WAYNE F. LEEBAW, LOUYSE A. LEE, AND PAUL D. WOOLF Endocrine-Metabolism Unit, Department of Medicine, University of Rochester Medical Center, Rochester, New York 14642 ABSTRACT. Although the role of the neurotransmitter, dopamine (DA), in the regulation of PRL has been well documented, controversy exists regarding its participation in the regulation of the other pituitary hormones. Consequently, we infused DA into six healthy male subjects (ages 19-32) and studied its effects on both basal pituitary hormone levels and augmented hormonal release induced by insulin hypoglycemia (ITT), TRH, and gonadotropin-releasing hormone (GnRH). DA alone produced a modest though significant increase in GH concentration from 2.2 ± 0.5 to 11.9 ± 3.7 ng/ml (P < 0.05) by 60 min, but the peak incremental GH response to ITT was significantly inhibited by DA (43.5 ± 5.0 us. 16.3 ± 3.3 ng/ml; P < 0.01). PRL concentrations fell during the DA infusion (20.4 ± 3.0 to 10.6 ± 1.5 ng/ml; P < 0.02) at 235 min, and the PRL responses to both ITT and TRH were completely abolished. Although the basal LH and FSH concentra-
T
HE ROLE of the neurotransmitter, dopamine (DA), in the regulation of many pituitary hormones has been well documented, especially for PRL secretion (1). DA and its analogues have also been shown to decrease basal TSH levels and to inhibit the TSH response to TRH (2), to reduce basal LH concentrations in normal women (3, 4), and to decrease FSH levels in women with hyperprolactinemic amenorrhea (4). Dopaminergic regulation of GH secretion is not well established. Although the DA precursor, L-dopa, stimulates GH release in patients with Parkinson's disease (5) and in normal humans (6), this effect may not be specific for
tions were unaffected by DA, the incremental LH response to GnRH was inhibited (45.5 ± 10.6 to 24.4 ± 5.4 mlU/ml; P < 0.05), while the FSH response was unchanged. DA significantly reduced the basal TSH concentration from 3.9 ± 0.2 to 2.5 ± 0.2 juU/ml (P < 0.01) at 230 min and blunted the peak incremental TSH response to TRH (6.0 ± 1.5 vs. 2.9 ± 0.9 juU/ml; P < 0.01). DA had no effect on basal cortisol levels, the cortisol response to ITT, basal plasma glucose, or the degree of hypoglycemia after ITT. Our data provide new evidence that DA has an inhibitory as well as a stimulatory role in the regulation of GH secretion in normal humans. It inhibits centrally as well as peripherally mediated PRL secretion and blunts the LH response to GnRH. In addition, DA lowers both basal and TRH-mediated TSH release, confirming the reports of other investigators. (J Clin Endocrinol Metab 47: 480, 1978)
DA (7-16). Dopaminergic inhibition of GH in -f acromegaly, however, has been clearly and repeatedly demonstrated (10, 11, 17-19), but this finding in pathological circumstances may have little relevance to normal physiology. The purpose of this study was to evaluate further dopaminergic effects on pituitary hormone release with emphasis on GH secretion. To achieve this, we infused DA itself, which does not cross the blood-brain barrier (20), and studied its effects on basal pituitary hormone levels and on augmented hormonal release induced by insulin hypoglycemia (ITT), TRH, and gonadotropin-releasing hormone (GnRH).
Materials and Methods Received September 20,1977. Address requests for reprints to: Dr. Paul D. Woolf, Assistant Professor of Medicine, Endocrine Metabolism Unit, Department of Medicine, University of Rochester Medical Center, 601 Elmwood Avenue, Rochester, New York 14642. * This study was supported in part by NINCDS Grant NS-11642, General Research Support Grant RR-05403, and USPHS Grant (NIH) RR-00044.
Six healthy male volunteers, 19-32 yr old, were chosen for this study. All were within 10% of ideal body weight, and none had a family history of diabetes or other endocrinopathy. Two iv insulin ~ tolerance tests were performed in random order on each subject: a control test and a test during DA infusion. TRH and GnRH were given together iv _
480
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 03:10 For personal use only. No other uses without permission. . All rights reserved.
DA EFFECT ON PITUITARY HORMONE SECRETION after the conclusion of each of the ITT's. In five of , the six subjects, a 345-min DA infusion without any provocative tests was performed. Each of these studies was separated by at least 1 week. The subjects were instructed to take nothing by mouth except water from 2300 h the night before • each study. After admission to the Clinical Research Center at 0730 h, 20-gauge catheters were inserted into each antecubital vein and kept open •, with a slow infusion (40 cc/h) of normal saline. One vein was used exclusively for infusion of saline or dopamine, while the other vein was used for drug administration and blood withdrawal. Blood pressure and pulse were taken every 15 min during the DA infusions. All studies began 1 h after catheter insertion. "The DA infusion was started 30 min before the , administration of insulin (0.1 U/kg iv) at a rate of 2 /ig/kg/min, increased at 15 min to 4 /ig/kg/min, .* and continued for a total of 345 min. One hundred and ninety-five minutes after insulin and 225 min into the DA infusion, a combination of TRH (500 _ /xg; Abbott Laboratories) and GnRH (100 ju,g; Parke, Davis, and Co.) was administered iv. Blood was obtained 30 and 15 min and immediately before insulin and at standard times after insulin, TRH, and GnRH. During the DA infusion alone, blood samples were obtained every 15 min for the 345 min of the infusion and for 45 min after its cessation. Blood samples were collected in heparinized tubes and centrifuged, and the plasma was frozen immediately after each study. Plasma glucose was , determined by the glucose oxidase method utilizing an Autoanalyzer. Plasma GH (21), TSH (22), LH (23), and FSH (24) were determined by RIA meth. ods previously described. Cortisol was determined by the method of Murphy (25). All values were the • means of duplicate determinations. The intra- and interassay coefficients of variation were: GH, 6.1% and 20.9%; TSH, 4.9% and 17.3%; LH, 6.4% and 12.1%; FSH, 3.1% and 11.9%; and cortisol, 10.2% and 10.8%. A new RIA for PRL was developed using a goat antihuman PRL sera generously given by Dr. Gerald Odstrchel of Corning Glass Works. The VLS-3 > PRL preparation provided by the National Pituitary Agency was used for the standards and was iodinated by the method of Greenwood, Hunter, and Glover (26). All reagents were diluted in 0.05 M phosphate buffer, pH 7.5, containing 0.1% sodium . azide and 1% bovine serum albumin (BSA). Each assay tube containing 100 /uJ standard or unknown, 100 jul antibody diluted 1:30,000, and 100 jul label (10,000 cpm) was incubated at room temperature for 24 h. Normal goat serum (100 \A) diluted 1:200
481
and antigoat y-globulin (100 jtil) made in the rabbit and diluted 1:10 were added before a second 24-h incubation also at room temperature. After the addition of 1 ml distilled water, each tube was centrifuged at 3000 rpm for 30 min, the supernatant was decanted, and the residual precipitate was counted. In the absence of standard (Bo), the antibody routinely bound 40-55% of the tracer, and the nonspecific binding was below 10%. In this system, as little PRL as 0.78 ng/ml (0.078 ng/tube) significantly decreased the percentage of label bound to the antibody and was the lowest standard used (Fig. 1). GH, LH, FSH, TSH, and insulin had no effect on the assay system (Fig. 1). The intraassay coefficient of variation (n = 20) at 12 ng/ml was 8.5% and at 65 ng/ml was 11.1%. The interassay coefficient of variation was 20.5% (n = 13). The PRL concentration in 25 normal men was 12.3 ± 1.0 ng/ml (SE) and was significantly lower than the value of 19.5 ± 1.5 ng/ml observed in 28 normal healthy women (P < 0.001). The normal range for men (mean ± 2 SD) was 2-22 and for women was 3-35 ng/ml. Statistical analysis was performed using the Student's t test for paired and unpaired data.
Results DA infusion had no untoward clinical effects, and the subjects noticed no subjective symptoms when the infusion was changed from saline to DA. Furthermore, compared to preinfusion values, DA had no effect on the vital signs (mean ± SEM; blood pressure, 119/70 ± 3.5/2.6 vs. 127/66 ± 6.6/4.8; pulse, 60.3 ± 2.3 vs. 62.4 ± 4.5, respectively). In contrast, during DA and ITT together, there was significant increases in the systolic blood pressure to 135 ± 5.0 (P < 0.025), decreases in
0
1.0
10
100
PROLACTIN (ng/hnl)
FIG. 1. The standard curve for the PRL assay. There is no cross-reactivity with LH (O O), FSH ( • •), TSH ( • • ) , human GH (X X), or insulin (V V). The minimum sensitivity of the assay is 0.78 ng/ml.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 03:10 For personal use only. No other uses without permission. . All rights reserved.
LEEBAW, LEE, AND WOOLF
482
1!..
JCE&M Vol 47
diastolic blood pressure to 54.5 ± 2.9 (P < 0.05), and quickening of the pulse rate compared to both the basal state and to DA alone. DA by itself did not affect the plasma glucose, nor did it alter the degree of hypoglycemia after insulin (33.8 ± 3.6 mg/dl vs. 37.5 ± 3.8 mg/dl during DA).
GH(Fig.2) DA alone produced a modest though significant increase in GH concentrations from 60-120 min (P < 0.05). The GH response that followed ITT alone was markedly blunted by DA with levels that were significantly lower than those observed during the saline infusion from 60 min (40.3 ± 5.2 vs. 15.6 ± 5.5 ng/ml; P < 0.005) after injection of insulin (90 min after the start of the DA) to the end of the test 120 min later (9.7 ± 2.9 vs. 2.3 ± 0.6 ng/ml; P < 0.05). At no time were the GH concentrations present during DA alone and during DA and ITT significantly different from each other. Furthermore, the maximal GH increment present during the combined DA and ITT (16.3 ± 3.3 ng/ml, mean ± SEM) was similar to those present during the DA infusion alone (11.2 ± 4.1 ng/ml), while both were significantly less than those present after insulin alone (43.4 ± 5.0 ng/ml; P < 0.01).
, 180
210
, 210
, 270
, 300
N
, 330
360
390
FIG. 3. Effects of DA on PRL secretion. DA alone diminished the basal PRL concentration (P < 0.05) at 60 min and abolished the PRL response to ITT and TRH. Insulin was injected at 30 min (thin arrow) and TRH (thick arrow) at 225 min into the DA infusion. Note the postDA rebound. The values represent the mean ± SEM. • • , DA alone; A- -A, ITT/TRH alone; O O, DA plus ITT/TRH.
3.0; P < 0.05), reached a nadir by 235 min (10.6 ± 1.5; P < 0.005), and remained significantly below basal values through the end of the infusion. Forty-five minutes after the DA infusion was stopped, PRL concentrations rebounded to 57.6 ± 18.9 ng/ml. The PRL responses to both ITT and TRH were completely abolished by the DA infusion. The baseline PRL levels were also significantly depressed. Gonadotropins
The basal LH concentration was unaffected by the infusion of DA alone. However, during DA, the LH response to GnRH (which was PRL (Fig. 3) given in both cases 195 min after insulin) was PRL concentrations fell significantly by 60 blunted, and the maximal LH increment was min of the DA infusion (20.4 ± 3.0 to 16.0 ± diminished from 45.5 ± 10.6 to 24.4 ± 5.4 mlU/ml (P < 0.05; Fig. 4, top panel). DA had no effect on basal FSH concentrations or the 50 FSH response to GnRH (Fig. 4, middle panel). ' )
40
I |
30
:S
20
TSH
-30
0
30
60
90
120
150
180
210
MINUTES
FIG. 2. Effects of DA on GH concentrations. DA alone resulted in a modest, significant increase in GH concentration by 60 min (P < 0.05) in five subjects, while it abolished the normal GH response to insulin-induced hypoglycemia (ITT) in six subjects. The insulin was given 30 min into the infusion (arrow). • • , DA alone; A- • -A, ITT alone; O---O, DA plus ITT. The values represent the mean ± SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.005 between ITT and DA plus ITT.
DA alone reduced the basal TSH concentration from 3.9 ± 0.2 juU/ml (mean of four basal values in the five subjects) to 2.5 ± 0.2 /xU/ml (P < 0.01) by 230 min with continued suppression throughout the infusion (Fig. 4, bottom panel). DA also inhibited the TSH response from 10-120 ijnin after TRH administration and significantly reduced the maximal incremental TSH response from 6.0 ±1.5 to 2.9 ± 0.9 juU/ml (P < 0.01; Fig. 4, bottom panel).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 03:10 For personal use only. No other uses without permission. . All rights reserved.
DA EFFECT ON PITUITARY HORMONE SECRETION
Cortisol A
The cortisol response to ITT also was unaffected by DA treatment (peak concentration, 23.5 ± 1.7 vs. 20.5 ± 1.4; maximal increment, 12.3 ± 1.3 vs. 11.0 ± 2.3 /xg/dl). There was no cortisol change during the infusion of DA (basal, 12.4 ± 2.4; range during infusion, 9.7 ± 2.4 to 14.9 ± 4.7 jug/dl), while the con-
n
60-, 5040-
\ I H
30-1
5
20-1 100-
1
*IK4^ Hh
16-
|
!
12
a,
i
4-
n rNS
012-,
n
10-
If::
8 >
-L. -1
c>
i
4-
•»
i
20Pre Oft
225
255
285
315
345
M
°x,
483
centrations during ITT and DA were significantly greater than those present during DA alone from 90-180 min after injection of insulin. Discussion Our results provide further insights into dopaminergic regulation of pituitary hormone secretion by demonstrating stimulation of basal GH release, inhibition of the augmented secretion of GH, PRL, LH, and TSH, and inhibition of basal PRL and TSH secretion. Our finding that DA is capable of abolishing the GH response to ITT has not previously been reported for DA or for any other dopaminergic agonist in normal humans. In fact, DA, though shown to markedly inhibit GH secretion in acromegaly (10, 19), has not been previously shown to have an inhibitory effect on GH levels in normal subjects and does not prevent the GH response to arginine (10). Indeed, preliminary observations from our laboratory indicate that DA may actually augment this response (unpublished observations). The inhibition by DA of the maximal GH response is more profound (80%) than the inhibition previously shown for the a-adrenergic antagonist, phentolamine (30-50%) (27), the serotonin antagonists, cyproheptadine (59%) and methysergide (35%) (28), or melatonin (45%) (29). The degree of hypoglycemia induced by insulin was unaffected by DA and was similar to that reported in these studies. The greater efficacy of DA might indicate a more important role for dopaminergic pathways than for noradrenergic or serotoningergic mechanisms in the mediation of this response. While the observations that L-dopa stimulates GH release (5, 6) support a dopaminergic mechanism for GH secretion, the inhibition of this response by phentolamine, an a-adrener-
MINUTES
FIG. 4. Effects of DA on LH, FSH, and TSH release. The LH response to GnRH (arrow) was blunted by DA, while basal levels were unaffected (top panel). (The pre-DA value was the mean of four samples in five subjects.) The mean incremental LH response to GnRH was significantly reduced by DA (ES). The basal FSH concentration and the FSH response to GnRH (arrow) were unchanged by DA (middle panel). Both the basal TSH concentration
and the TSH response to TRH (arrow) and were significantly depressed by DA (bottompanel). DA significantly reduced the incremental TSH response to TRH (s). • • , DA alone; A- • -A, releasing hormone; O O, DA plus releasing hormone. The values represent the mean ± SEM. *, P < 0.05; •*, P < 0.01; ***, P < 0.005 between releasing hormone alone and releasing hormone plus DA.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 03:10 For personal use only. No other uses without permission. . All rights reserved.
484
LEEBAW, LEE, AND WOOLF
JCE&M Vol47
1978 No 3
HYPOGLYCEMIA gic blocking agent (7), and cyproheptadine (8), Ia serotoninergic blocking agent, indicates that HYP L-DOFA-*this effect may be mediated by other neurotransmitters. The dopamine agonist, bromoerBBB I gocryptine, causes only slight stimulation of GH release (11), while apomorphine, another ME agonist (12), consistently stimulates GH secretion (13, 14). The recent demonstration of PIT blockade of apomorphine-induced GH secretion by pimozide, a specific dopamine antagonist, lends further support for dopaminergic regulation (15). However, apomorphine also FIG. 5. Schematic representation of the effects of DA on 4 GH release. Using Martin's proposed mechanism for dohas been shown to affect serotoninergic mech- paminergic regulation of GH secretion (31), our data anisms. In our hands, DA itself caused a mod- indicate that DA has weak stimulatory and potent inhibest but significant rise in GH levels with four itory effects at the axo-axonal synapse of the dopami- * of the five subjects having GH increments >5 nergic neuron with the peptidergic GRF neuron within
ng/ml, confirming the observations of some
investigators (9), but not of others (10, 19). Although this response may be caused by nonspecific stress, this is unlikely in our study because the infusion of DA was begun 1 h after catheter insertion, at which time mean GH levels were below 3 ng/ml and cardiovascular changes were not apparent. The transient, significant rise in basal GH concentration and blockade of hypoglycemiainduced GH release suggests a dual effect of DA on GH secretion; i.e. stimulation of basal GH release and inhibition of augmented GH secretion. This is analogous to the effects of melatonin, which by itself stimulates GH secretion (30) while blunting the hypoglycemiamediated GH release (29). Although the mechanism of this dual effect is unknown and almost certainly complex, we propose the following hypothesis as a possible explanation. It has been postulated that hypothalamic dopaminergic neurons stimulate peptidergic GH-releasing factor (GRF) neurons via axo-axonal synapses at the level of the median eminence (31) (Fig. 5). Therefore, stimuli that increase endogenous central nervous system DA, such as L-dopa, result in release of GRF and, hence, of GH. It follows from this concept that the administration of exogenous DA, which does not enter the brain (20), should also directly facilitate peptidergic GRF axonal release, as these axons are located within the median eminence (ME) which lies outside the blood-brain barrier (Fig. 5). This
the median eminence. BBB, Blood-brain barrier; HYP, hypothalamus; Pit, pituitary. t
DA effect may be dose related (lower concentrations, which would be expected to be pres- < ent early in the infusion because of the lower initial infusion rate, are stimulatory, while higher concentrations are inhibitory), similar />' to the preliminary observations reported for the effects of the DA agonist, apomorphine, on GH release in the rat (32). Unfortunately, -• DA levels were not determined. Alternatively, the continuing administration of DA may prevent hypoglycemia-induced GRF release by acting as a weak agonist/antagonist on the peptidergic GRF axon, or by negative feed- < back inhibition of endogenous DA formation or release within hypothalamic dopaminergic neurons (33,34). In either case, any facilitation of GRF secretion and, hence, of GH release by ITT that depended upon dopaminergic transmission at the ME would be effectively * blocked. This theory for the peripheral action of DA is consistent with the well known stimulation of basal GH secretion induced by the centrally acting dopaminergic drugs, L-dopa . (5, 6) and apomorphine (13, 14). However, central stimulation seems to be more potent and longer lasting than that induced peripherally by DA itself. Whether centrally acting dopaminergic drugs also inhibit stimulated GH secretion is unknown. Alternatively, DA from either the peripheral or portal circulation may act directly on A"
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 03:10 For personal use only. No other uses without permission. . All rights reserved.
DA EFFECT ON PITUITARY HORMONE SECRETION the pituitary, causing weak stimulation of GH while antagonizing the effects of endogenous
Vol. 47, No. 3 Printed in U.S.A.
Dopamine Affects Basal and Augmented Pituitary Hormone Secretion* WAYNE F. LEEBAW, LOUYSE A. LEE, AND PAUL D. WOOLF Endocrine-Metabolism Unit, Department of Medicine, University of Rochester Medical Center, Rochester, New York 14642 ABSTRACT. Although the role of the neurotransmitter, dopamine (DA), in the regulation of PRL has been well documented, controversy exists regarding its participation in the regulation of the other pituitary hormones. Consequently, we infused DA into six healthy male subjects (ages 19-32) and studied its effects on both basal pituitary hormone levels and augmented hormonal release induced by insulin hypoglycemia (ITT), TRH, and gonadotropin-releasing hormone (GnRH). DA alone produced a modest though significant increase in GH concentration from 2.2 ± 0.5 to 11.9 ± 3.7 ng/ml (P < 0.05) by 60 min, but the peak incremental GH response to ITT was significantly inhibited by DA (43.5 ± 5.0 us. 16.3 ± 3.3 ng/ml; P < 0.01). PRL concentrations fell during the DA infusion (20.4 ± 3.0 to 10.6 ± 1.5 ng/ml; P < 0.02) at 235 min, and the PRL responses to both ITT and TRH were completely abolished. Although the basal LH and FSH concentra-
T
HE ROLE of the neurotransmitter, dopamine (DA), in the regulation of many pituitary hormones has been well documented, especially for PRL secretion (1). DA and its analogues have also been shown to decrease basal TSH levels and to inhibit the TSH response to TRH (2), to reduce basal LH concentrations in normal women (3, 4), and to decrease FSH levels in women with hyperprolactinemic amenorrhea (4). Dopaminergic regulation of GH secretion is not well established. Although the DA precursor, L-dopa, stimulates GH release in patients with Parkinson's disease (5) and in normal humans (6), this effect may not be specific for
tions were unaffected by DA, the incremental LH response to GnRH was inhibited (45.5 ± 10.6 to 24.4 ± 5.4 mlU/ml; P < 0.05), while the FSH response was unchanged. DA significantly reduced the basal TSH concentration from 3.9 ± 0.2 to 2.5 ± 0.2 juU/ml (P < 0.01) at 230 min and blunted the peak incremental TSH response to TRH (6.0 ± 1.5 vs. 2.9 ± 0.9 juU/ml; P < 0.01). DA had no effect on basal cortisol levels, the cortisol response to ITT, basal plasma glucose, or the degree of hypoglycemia after ITT. Our data provide new evidence that DA has an inhibitory as well as a stimulatory role in the regulation of GH secretion in normal humans. It inhibits centrally as well as peripherally mediated PRL secretion and blunts the LH response to GnRH. In addition, DA lowers both basal and TRH-mediated TSH release, confirming the reports of other investigators. (J Clin Endocrinol Metab 47: 480, 1978)
DA (7-16). Dopaminergic inhibition of GH in -f acromegaly, however, has been clearly and repeatedly demonstrated (10, 11, 17-19), but this finding in pathological circumstances may have little relevance to normal physiology. The purpose of this study was to evaluate further dopaminergic effects on pituitary hormone release with emphasis on GH secretion. To achieve this, we infused DA itself, which does not cross the blood-brain barrier (20), and studied its effects on basal pituitary hormone levels and on augmented hormonal release induced by insulin hypoglycemia (ITT), TRH, and gonadotropin-releasing hormone (GnRH).
Materials and Methods Received September 20,1977. Address requests for reprints to: Dr. Paul D. Woolf, Assistant Professor of Medicine, Endocrine Metabolism Unit, Department of Medicine, University of Rochester Medical Center, 601 Elmwood Avenue, Rochester, New York 14642. * This study was supported in part by NINCDS Grant NS-11642, General Research Support Grant RR-05403, and USPHS Grant (NIH) RR-00044.
Six healthy male volunteers, 19-32 yr old, were chosen for this study. All were within 10% of ideal body weight, and none had a family history of diabetes or other endocrinopathy. Two iv insulin ~ tolerance tests were performed in random order on each subject: a control test and a test during DA infusion. TRH and GnRH were given together iv _
480
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 03:10 For personal use only. No other uses without permission. . All rights reserved.
DA EFFECT ON PITUITARY HORMONE SECRETION after the conclusion of each of the ITT's. In five of , the six subjects, a 345-min DA infusion without any provocative tests was performed. Each of these studies was separated by at least 1 week. The subjects were instructed to take nothing by mouth except water from 2300 h the night before • each study. After admission to the Clinical Research Center at 0730 h, 20-gauge catheters were inserted into each antecubital vein and kept open •, with a slow infusion (40 cc/h) of normal saline. One vein was used exclusively for infusion of saline or dopamine, while the other vein was used for drug administration and blood withdrawal. Blood pressure and pulse were taken every 15 min during the DA infusions. All studies began 1 h after catheter insertion. "The DA infusion was started 30 min before the , administration of insulin (0.1 U/kg iv) at a rate of 2 /ig/kg/min, increased at 15 min to 4 /ig/kg/min, .* and continued for a total of 345 min. One hundred and ninety-five minutes after insulin and 225 min into the DA infusion, a combination of TRH (500 _ /xg; Abbott Laboratories) and GnRH (100 ju,g; Parke, Davis, and Co.) was administered iv. Blood was obtained 30 and 15 min and immediately before insulin and at standard times after insulin, TRH, and GnRH. During the DA infusion alone, blood samples were obtained every 15 min for the 345 min of the infusion and for 45 min after its cessation. Blood samples were collected in heparinized tubes and centrifuged, and the plasma was frozen immediately after each study. Plasma glucose was , determined by the glucose oxidase method utilizing an Autoanalyzer. Plasma GH (21), TSH (22), LH (23), and FSH (24) were determined by RIA meth. ods previously described. Cortisol was determined by the method of Murphy (25). All values were the • means of duplicate determinations. The intra- and interassay coefficients of variation were: GH, 6.1% and 20.9%; TSH, 4.9% and 17.3%; LH, 6.4% and 12.1%; FSH, 3.1% and 11.9%; and cortisol, 10.2% and 10.8%. A new RIA for PRL was developed using a goat antihuman PRL sera generously given by Dr. Gerald Odstrchel of Corning Glass Works. The VLS-3 > PRL preparation provided by the National Pituitary Agency was used for the standards and was iodinated by the method of Greenwood, Hunter, and Glover (26). All reagents were diluted in 0.05 M phosphate buffer, pH 7.5, containing 0.1% sodium . azide and 1% bovine serum albumin (BSA). Each assay tube containing 100 /uJ standard or unknown, 100 jul antibody diluted 1:30,000, and 100 jul label (10,000 cpm) was incubated at room temperature for 24 h. Normal goat serum (100 \A) diluted 1:200
481
and antigoat y-globulin (100 jtil) made in the rabbit and diluted 1:10 were added before a second 24-h incubation also at room temperature. After the addition of 1 ml distilled water, each tube was centrifuged at 3000 rpm for 30 min, the supernatant was decanted, and the residual precipitate was counted. In the absence of standard (Bo), the antibody routinely bound 40-55% of the tracer, and the nonspecific binding was below 10%. In this system, as little PRL as 0.78 ng/ml (0.078 ng/tube) significantly decreased the percentage of label bound to the antibody and was the lowest standard used (Fig. 1). GH, LH, FSH, TSH, and insulin had no effect on the assay system (Fig. 1). The intraassay coefficient of variation (n = 20) at 12 ng/ml was 8.5% and at 65 ng/ml was 11.1%. The interassay coefficient of variation was 20.5% (n = 13). The PRL concentration in 25 normal men was 12.3 ± 1.0 ng/ml (SE) and was significantly lower than the value of 19.5 ± 1.5 ng/ml observed in 28 normal healthy women (P < 0.001). The normal range for men (mean ± 2 SD) was 2-22 and for women was 3-35 ng/ml. Statistical analysis was performed using the Student's t test for paired and unpaired data.
Results DA infusion had no untoward clinical effects, and the subjects noticed no subjective symptoms when the infusion was changed from saline to DA. Furthermore, compared to preinfusion values, DA had no effect on the vital signs (mean ± SEM; blood pressure, 119/70 ± 3.5/2.6 vs. 127/66 ± 6.6/4.8; pulse, 60.3 ± 2.3 vs. 62.4 ± 4.5, respectively). In contrast, during DA and ITT together, there was significant increases in the systolic blood pressure to 135 ± 5.0 (P < 0.025), decreases in
0
1.0
10
100
PROLACTIN (ng/hnl)
FIG. 1. The standard curve for the PRL assay. There is no cross-reactivity with LH (O O), FSH ( • •), TSH ( • • ) , human GH (X X), or insulin (V V). The minimum sensitivity of the assay is 0.78 ng/ml.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 03:10 For personal use only. No other uses without permission. . All rights reserved.
LEEBAW, LEE, AND WOOLF
482
1!..
JCE&M Vol 47
diastolic blood pressure to 54.5 ± 2.9 (P < 0.05), and quickening of the pulse rate compared to both the basal state and to DA alone. DA by itself did not affect the plasma glucose, nor did it alter the degree of hypoglycemia after insulin (33.8 ± 3.6 mg/dl vs. 37.5 ± 3.8 mg/dl during DA).
GH(Fig.2) DA alone produced a modest though significant increase in GH concentrations from 60-120 min (P < 0.05). The GH response that followed ITT alone was markedly blunted by DA with levels that were significantly lower than those observed during the saline infusion from 60 min (40.3 ± 5.2 vs. 15.6 ± 5.5 ng/ml; P < 0.005) after injection of insulin (90 min after the start of the DA) to the end of the test 120 min later (9.7 ± 2.9 vs. 2.3 ± 0.6 ng/ml; P < 0.05). At no time were the GH concentrations present during DA alone and during DA and ITT significantly different from each other. Furthermore, the maximal GH increment present during the combined DA and ITT (16.3 ± 3.3 ng/ml, mean ± SEM) was similar to those present during the DA infusion alone (11.2 ± 4.1 ng/ml), while both were significantly less than those present after insulin alone (43.4 ± 5.0 ng/ml; P < 0.01).
, 180
210
, 210
, 270
, 300
N
, 330
360
390
FIG. 3. Effects of DA on PRL secretion. DA alone diminished the basal PRL concentration (P < 0.05) at 60 min and abolished the PRL response to ITT and TRH. Insulin was injected at 30 min (thin arrow) and TRH (thick arrow) at 225 min into the DA infusion. Note the postDA rebound. The values represent the mean ± SEM. • • , DA alone; A- -A, ITT/TRH alone; O O, DA plus ITT/TRH.
3.0; P < 0.05), reached a nadir by 235 min (10.6 ± 1.5; P < 0.005), and remained significantly below basal values through the end of the infusion. Forty-five minutes after the DA infusion was stopped, PRL concentrations rebounded to 57.6 ± 18.9 ng/ml. The PRL responses to both ITT and TRH were completely abolished by the DA infusion. The baseline PRL levels were also significantly depressed. Gonadotropins
The basal LH concentration was unaffected by the infusion of DA alone. However, during DA, the LH response to GnRH (which was PRL (Fig. 3) given in both cases 195 min after insulin) was PRL concentrations fell significantly by 60 blunted, and the maximal LH increment was min of the DA infusion (20.4 ± 3.0 to 16.0 ± diminished from 45.5 ± 10.6 to 24.4 ± 5.4 mlU/ml (P < 0.05; Fig. 4, top panel). DA had no effect on basal FSH concentrations or the 50 FSH response to GnRH (Fig. 4, middle panel). ' )
40
I |
30
:S
20
TSH
-30
0
30
60
90
120
150
180
210
MINUTES
FIG. 2. Effects of DA on GH concentrations. DA alone resulted in a modest, significant increase in GH concentration by 60 min (P < 0.05) in five subjects, while it abolished the normal GH response to insulin-induced hypoglycemia (ITT) in six subjects. The insulin was given 30 min into the infusion (arrow). • • , DA alone; A- • -A, ITT alone; O---O, DA plus ITT. The values represent the mean ± SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.005 between ITT and DA plus ITT.
DA alone reduced the basal TSH concentration from 3.9 ± 0.2 juU/ml (mean of four basal values in the five subjects) to 2.5 ± 0.2 /xU/ml (P < 0.01) by 230 min with continued suppression throughout the infusion (Fig. 4, bottom panel). DA also inhibited the TSH response from 10-120 ijnin after TRH administration and significantly reduced the maximal incremental TSH response from 6.0 ±1.5 to 2.9 ± 0.9 juU/ml (P < 0.01; Fig. 4, bottom panel).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 03:10 For personal use only. No other uses without permission. . All rights reserved.
DA EFFECT ON PITUITARY HORMONE SECRETION
Cortisol A
The cortisol response to ITT also was unaffected by DA treatment (peak concentration, 23.5 ± 1.7 vs. 20.5 ± 1.4; maximal increment, 12.3 ± 1.3 vs. 11.0 ± 2.3 /xg/dl). There was no cortisol change during the infusion of DA (basal, 12.4 ± 2.4; range during infusion, 9.7 ± 2.4 to 14.9 ± 4.7 jug/dl), while the con-
n
60-, 5040-
\ I H
30-1
5
20-1 100-
1
*IK4^ Hh
16-
|
!
12
a,
i
4-
n rNS
012-,
n
10-
If::
8 >
-L. -1
c>
i
4-
•»
i
20Pre Oft
225
255
285
315
345
M
°x,
483
centrations during ITT and DA were significantly greater than those present during DA alone from 90-180 min after injection of insulin. Discussion Our results provide further insights into dopaminergic regulation of pituitary hormone secretion by demonstrating stimulation of basal GH release, inhibition of the augmented secretion of GH, PRL, LH, and TSH, and inhibition of basal PRL and TSH secretion. Our finding that DA is capable of abolishing the GH response to ITT has not previously been reported for DA or for any other dopaminergic agonist in normal humans. In fact, DA, though shown to markedly inhibit GH secretion in acromegaly (10, 19), has not been previously shown to have an inhibitory effect on GH levels in normal subjects and does not prevent the GH response to arginine (10). Indeed, preliminary observations from our laboratory indicate that DA may actually augment this response (unpublished observations). The inhibition by DA of the maximal GH response is more profound (80%) than the inhibition previously shown for the a-adrenergic antagonist, phentolamine (30-50%) (27), the serotonin antagonists, cyproheptadine (59%) and methysergide (35%) (28), or melatonin (45%) (29). The degree of hypoglycemia induced by insulin was unaffected by DA and was similar to that reported in these studies. The greater efficacy of DA might indicate a more important role for dopaminergic pathways than for noradrenergic or serotoningergic mechanisms in the mediation of this response. While the observations that L-dopa stimulates GH release (5, 6) support a dopaminergic mechanism for GH secretion, the inhibition of this response by phentolamine, an a-adrener-
MINUTES
FIG. 4. Effects of DA on LH, FSH, and TSH release. The LH response to GnRH (arrow) was blunted by DA, while basal levels were unaffected (top panel). (The pre-DA value was the mean of four samples in five subjects.) The mean incremental LH response to GnRH was significantly reduced by DA (ES). The basal FSH concentration and the FSH response to GnRH (arrow) were unchanged by DA (middle panel). Both the basal TSH concentration
and the TSH response to TRH (arrow) and were significantly depressed by DA (bottompanel). DA significantly reduced the incremental TSH response to TRH (s). • • , DA alone; A- • -A, releasing hormone; O O, DA plus releasing hormone. The values represent the mean ± SEM. *, P < 0.05; •*, P < 0.01; ***, P < 0.005 between releasing hormone alone and releasing hormone plus DA.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 03:10 For personal use only. No other uses without permission. . All rights reserved.
484
LEEBAW, LEE, AND WOOLF
JCE&M Vol47
1978 No 3
HYPOGLYCEMIA gic blocking agent (7), and cyproheptadine (8), Ia serotoninergic blocking agent, indicates that HYP L-DOFA-*this effect may be mediated by other neurotransmitters. The dopamine agonist, bromoerBBB I gocryptine, causes only slight stimulation of GH release (11), while apomorphine, another ME agonist (12), consistently stimulates GH secretion (13, 14). The recent demonstration of PIT blockade of apomorphine-induced GH secretion by pimozide, a specific dopamine antagonist, lends further support for dopaminergic regulation (15). However, apomorphine also FIG. 5. Schematic representation of the effects of DA on 4 GH release. Using Martin's proposed mechanism for dohas been shown to affect serotoninergic mech- paminergic regulation of GH secretion (31), our data anisms. In our hands, DA itself caused a mod- indicate that DA has weak stimulatory and potent inhibest but significant rise in GH levels with four itory effects at the axo-axonal synapse of the dopami- * of the five subjects having GH increments >5 nergic neuron with the peptidergic GRF neuron within
ng/ml, confirming the observations of some
investigators (9), but not of others (10, 19). Although this response may be caused by nonspecific stress, this is unlikely in our study because the infusion of DA was begun 1 h after catheter insertion, at which time mean GH levels were below 3 ng/ml and cardiovascular changes were not apparent. The transient, significant rise in basal GH concentration and blockade of hypoglycemiainduced GH release suggests a dual effect of DA on GH secretion; i.e. stimulation of basal GH release and inhibition of augmented GH secretion. This is analogous to the effects of melatonin, which by itself stimulates GH secretion (30) while blunting the hypoglycemiamediated GH release (29). Although the mechanism of this dual effect is unknown and almost certainly complex, we propose the following hypothesis as a possible explanation. It has been postulated that hypothalamic dopaminergic neurons stimulate peptidergic GH-releasing factor (GRF) neurons via axo-axonal synapses at the level of the median eminence (31) (Fig. 5). Therefore, stimuli that increase endogenous central nervous system DA, such as L-dopa, result in release of GRF and, hence, of GH. It follows from this concept that the administration of exogenous DA, which does not enter the brain (20), should also directly facilitate peptidergic GRF axonal release, as these axons are located within the median eminence (ME) which lies outside the blood-brain barrier (Fig. 5). This
the median eminence. BBB, Blood-brain barrier; HYP, hypothalamus; Pit, pituitary. t
DA effect may be dose related (lower concentrations, which would be expected to be pres- < ent early in the infusion because of the lower initial infusion rate, are stimulatory, while higher concentrations are inhibitory), similar />' to the preliminary observations reported for the effects of the DA agonist, apomorphine, on GH release in the rat (32). Unfortunately, -• DA levels were not determined. Alternatively, the continuing administration of DA may prevent hypoglycemia-induced GRF release by acting as a weak agonist/antagonist on the peptidergic GRF axon, or by negative feed- < back inhibition of endogenous DA formation or release within hypothalamic dopaminergic neurons (33,34). In either case, any facilitation of GRF secretion and, hence, of GH release by ITT that depended upon dopaminergic transmission at the ME would be effectively * blocked. This theory for the peripheral action of DA is consistent with the well known stimulation of basal GH secretion induced by the centrally acting dopaminergic drugs, L-dopa . (5, 6) and apomorphine (13, 14). However, central stimulation seems to be more potent and longer lasting than that induced peripherally by DA itself. Whether centrally acting dopaminergic drugs also inhibit stimulated GH secretion is unknown. Alternatively, DA from either the peripheral or portal circulation may act directly on A"
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 November 2015. at 03:10 For personal use only. No other uses without permission. . All rights reserved.
DA EFFECT ON PITUITARY HORMONE SECRETION the pituitary, causing weak stimulation of GH while antagonizing the effects of endogenous
