0021-972X/91/7205-1081$03.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1991 by The Endocrine Society
Vol. 72, No. 5 Printed in U.S.A.
The Role of the Cholinergic Pathway in Growth Hormone Feedback* MIRTHA KELIJMAN AND LAWRENCE A. FROHMAN Division of Endocrinology and Metabolism, Department of Internal Medicine, University of Cincinnati College of Medicine, Cincinnati, Ohio 45267
ABSTRACT. Cholinergic pathways play an important role in the regulation of GH secretion. To assess their participation in GH feedback, we investigated the effect of pyridostigmine (an acetylcholinesterase inhibitor) on plasma GH responses to GHreleasing hormone (GHRH) plus TRH, insulin hypoglycemia, and arginine as well as on the inhibition of these responses by exogenous GH. The GH response to each stimulus was inhibited by an infusion of GH (0.55 Mg/mVmin)» started 4 h earlier. Pyridostigmine (120 mg, orally), administered 30 min before the stimulus, enhanced GH responses to GHRH and insulin during both saline and GH infusions. However, GH responses during combined administration of pyridostigmine and GH were less than those during pyridostigmine alone. GH responses to argi-
nine, in contrast, were not affected by pyridostigmine in either the absence or presence of exogenous GH. TSH responses to TRH were unaltered by either GH or pyridostigmine. Pyridostigmine enhancement of GH responses to a maximally stimulatory dose of GHRH suggests that its effect is exerted by inhibition of somatostatin release. The lack of effect of pyridostigmine on plasma GH responses to arginine suggests that arginine and pyridostigmine increase GH secretion through a common pathway. The enhancement by pyridostigmine of GH responses in both the presence and absence of exogenous GH suggests that exogenous GH and pyridostigmine exert their discordant effects on GH secretion through independent mechanisms. (J Clin Endocrinol Metab 72: 1081-1087, 1991)
T
secretion (16-18). In the present study we have investigated whether direct GH feedback is mediated by a cholinergic pathway by determining whether the inhibition of GH secretion induced by short term GH infusion is prevented by increasing cholinergic tone by pyridostigmine, an acetylcholinesterase inhibitor. In an attempt to determine whether cholinergic modulation of GH feedback occurs in the pituitary or the hypothalamus, we studied the cholinergic modulation of GH feedback on the responses to pituitary GHRH and hypothalamicmediated (insulin hypoglycemia and arginine) stimuli of GH secretion.
HE SECRETION of GH is inhibited by previous GH administration in both men (1-4) and experimental animals (5-7). This negative feedback is probably effected both directly by GH and indirectly through GHdependent insulin-like growth factor-I (IGF-I). While the latter inhibits GH secretion by a dual effect on both the pituitary and the hypothalamus (8), no direct effect of GH on the somatotroph has been observed in in vitro studies (9, 10). The absence of a direct effect of GH on the pituitary indicates that the inhibition of GH secretion observed after GH administration without elevation of plasma IGF-I levels (11,12) is mediated by the central nervous system (CNS). Acetylcholine modulation of GH secretion in man has been extensively documented. Several studies have shown that cholinergic blockade inhibits spontaneous and exercise-, GH-releasing hormone (GHRH)-, insulin hypoglycemia-, arginine-, and clonidine-stimulated GH secretion (13-15). Conversely, enhanced cholinergic tone induced by acetylcholinesterase inhibitors augments spontaneous and GHRH- and clonidine-stimulated GH
Materials and Methods Subjects
Received August 17, 1990. Address all correspondence and requests for reprints to: Lawrence A. Frohman, M.D., Division of Endocrinology and Metabolism, University of Cincinnati College of Medicine, 231 Bethesda Avenue, ML 547, Cincinnati, Ohio 45267. * This work was supported in part by USPHS Grants DK-30667 and RR-00068 and a grant from Genentech, Inc.
Sixteen men, aged 18-34 yr (mean ± SE, 23.6 ± 1.1) participated in the study. Some of the subjects had participated in a previous research protocol (12) that included studies under conditions similar to those of studies in the present research protocol; data obtained at that time were included in the present study. All subjects were within 15% of ideal body weight and were receiving no medications. Each subject gave written informed consent to the protocols, which were approved by the Institutional Review Board. Protocols were performed in the University of Cincinnati General Clinical Research Center, with an interval of at least 7 days between studies. The order of the studies was randomly assigned. The subjects fasted from
1081
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 November 2015. at 13:32 For personal use only. No other uses without permission. . All rights reserved.
KELIJMAN AND FROHMAN
1082
midnight until completion of the study. Indwelling needles were inserted in both forearm veins between 0730-0830 h, and subjects remained recumbent and awake during the studies. Protocols Combined GHRH and TRH stimulation Six men received 6-h infusions of biosynthetic GH (Protropin, Genentech, Inc., South San Francisco, CA) at a rate (0.55 Mg/m2/min) we have previously shown to significantly inhibit plasma GH responses to GHRH stimulation (12) or 150 mmol/ L NaCl. Four hours after the initiation of the infusion, GHRH(1-44) (1 Mg/kg; Hoffmann-LaRoche, Inc., Nutley, NJ) and TRH (0.3 jug/kg; Thypinone, Abbott Laboratories, North Chicago, IL) were injected iv. A maximally stimulatory dose of GHRH was chosen to exclude the effects of endogenous GHRH secretion. These studies were repeated with the addition of 120 mg pyridostigmine administered orally 30 min before the GHRH/TRH injection. It has been previously shown that administration of this dose of pyridostigmine results in enhanced plasma GH responses (17). GH, GHRH, and TRH were diluted as previously described (12). Insulin hypoglycemia and arginine stimulation In similar protocols, regular insulin (0.1 U/kg, iv bolus injection; six subjects) or L-arginine hydrochloride (30 g infused iv over a 30-min period; six subjects) was substituted for GHRH plus TRH as the stimulus for hormone secretion. Blood glucose levels were measured at 15-min intervals after insulin injection.
JCE&M«1991 Vol 72 • No 5
Statistical analysis Plasma GH responses to GHRH, TRH, insulin, and arginine were quantified by measurement of peak incremental plasma hormone levels and integrated secretion, as determined by the trapezoidal method. Basal GH concentrations were defined as the mean of the plasma GH values during the 2-h period preceding the stimulus. Plasma GH, IGF-I, glucose, and FFA levels before stimuli administration in all protocols were pooled. Group differences were analyzed by a repeated measures twofactor analysis of variance (SAS Institute, Cary, NC) for multiple comparisons after logarithmic transformation when indicated. A one-tailed P value of less than 0.05 for effects of exogenous GH and pyridostigmine was considered significant.
Results Effects of exogenous GH and pyridostigmine on GH and TSH responses to GHRH and TRH Plasma GH responses and the integrated secretory response to GHRH plus TRH are shown in Fig. 1 and Table 1. Exogenous GH significantly inhibited GH secretory responses, expressed as both peak (41%; P = 0.01) and integrated (48%; P < 0.01) GH secretion. In contrast, pyridostigmine significantly enhanced both 40
GH (0.55 ug/m7min) or SAUNE PYRIDOSTIGMINE GHRH + TRH
30
Sample collection and processing Blood samples were collected in heparinized tubes every 15 min throughout each study for measurement of plasma GH, TSH, and PRL, as indicated below. Samples for glucose, IGFI, and FFA determinations were obtained immediately before the saline/GH infusion and immediately before GHRH plus TRH or insulin injection or arginine infusion. All blood samples were immediately chilled, centrifuged at 4 C, and stored at -20 C until assayed. Plasma hormone concentrations in each study were determined in a single assay. Plasma GH, TSH, and PRL were measured in duplicate by RIA, as previously described (19-21), with intra- and interassay coefficients of variation of 6.1% and 6.8%, 7.2% and 7.8%, 10.2% and 10.7%, and 4.5% and 9.1%, respectively. Plasma IGF-I was assayed by a modification of the method of Daughaday et al. (22); biosynthetic IGF-I (Amgen Biologicals, Thousand Oaks, CA) was used for iodination and as standard. AntiIGF-I serum was provided by the National Hormone and Pituitary Program, NIDDK (Baltimore, MD). The intra- and interassay coefficients of variation were 4.1% and 10.5%, respectively. Plasma glucose levels were measured by a glucose oxidase method. Serum FFA concentrations were determined by an enzymatic colorimetric method (Wako Pure Chemical Industries, Ltd., Osaka, Japan).
\
20
o.-o CONTROL (6)
10
a--* GH(7) •— PYR(6) ^
30
GH + PYR (6)
240
270
300 MINUTES
330
360
390
FIG. 1. Effects of exogenous GH and pyridostigmine (PYR), alone and in combination, on the plasma GH response (A) to GRH (1 fig/kg, iv). Shown are the mean ± SE. The number of subjects in each group is shown in parentheses. Baseline plasma GH levels were 0.4 ± 0.04 Mg/L (control), 3.6 ± 0.1 Mg/L (GH infusion), 0.4 ± 0.03 Mg/L (pyridostigmine), and 4.0 ± 0.4 ^g/L (GH infusion plus pyridostigmine). The increase in basal GH levels attributable to the infusion of exogenous GH was subtracted from the measured values to facilitate comparison of the responses to stimuli.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 November 2015. at 13:32 For personal use only. No other uses without permission. . All rights reserved.
CHOLINERGIC MODULATION OF GH FEEDBACK
1083
TABLE 1. Effect of exogenous GH and pyridostigmine on plasma GH responses
Stimulus GHRH (1 Mg/kg) No. of subjects Peak increment iixg/h) Integrated response (^g/L • min) Insulin (0.1 U/kg) No. of subjects Peak increment (ng/L) Integrated response (^g/Lmin) Arginine (30 g) No. of subjects Peak increment (^g/L) Integrated response (fig/L • min)
Combined effects0
Control
GH
Pyridostigmine
GH + pyridostigmine
6 21.6 ± 5.0 1434 ± 332
7 3.6 ± 0.1 353 ± 47
6 29.8 ± 7.2 2436 ± 580
6 25.5 ± 5.4 1761 ± 406
0.01
Vol. 72, No. 5 Printed in U.S.A.
The Role of the Cholinergic Pathway in Growth Hormone Feedback* MIRTHA KELIJMAN AND LAWRENCE A. FROHMAN Division of Endocrinology and Metabolism, Department of Internal Medicine, University of Cincinnati College of Medicine, Cincinnati, Ohio 45267
ABSTRACT. Cholinergic pathways play an important role in the regulation of GH secretion. To assess their participation in GH feedback, we investigated the effect of pyridostigmine (an acetylcholinesterase inhibitor) on plasma GH responses to GHreleasing hormone (GHRH) plus TRH, insulin hypoglycemia, and arginine as well as on the inhibition of these responses by exogenous GH. The GH response to each stimulus was inhibited by an infusion of GH (0.55 Mg/mVmin)» started 4 h earlier. Pyridostigmine (120 mg, orally), administered 30 min before the stimulus, enhanced GH responses to GHRH and insulin during both saline and GH infusions. However, GH responses during combined administration of pyridostigmine and GH were less than those during pyridostigmine alone. GH responses to argi-
nine, in contrast, were not affected by pyridostigmine in either the absence or presence of exogenous GH. TSH responses to TRH were unaltered by either GH or pyridostigmine. Pyridostigmine enhancement of GH responses to a maximally stimulatory dose of GHRH suggests that its effect is exerted by inhibition of somatostatin release. The lack of effect of pyridostigmine on plasma GH responses to arginine suggests that arginine and pyridostigmine increase GH secretion through a common pathway. The enhancement by pyridostigmine of GH responses in both the presence and absence of exogenous GH suggests that exogenous GH and pyridostigmine exert their discordant effects on GH secretion through independent mechanisms. (J Clin Endocrinol Metab 72: 1081-1087, 1991)
T
secretion (16-18). In the present study we have investigated whether direct GH feedback is mediated by a cholinergic pathway by determining whether the inhibition of GH secretion induced by short term GH infusion is prevented by increasing cholinergic tone by pyridostigmine, an acetylcholinesterase inhibitor. In an attempt to determine whether cholinergic modulation of GH feedback occurs in the pituitary or the hypothalamus, we studied the cholinergic modulation of GH feedback on the responses to pituitary GHRH and hypothalamicmediated (insulin hypoglycemia and arginine) stimuli of GH secretion.
HE SECRETION of GH is inhibited by previous GH administration in both men (1-4) and experimental animals (5-7). This negative feedback is probably effected both directly by GH and indirectly through GHdependent insulin-like growth factor-I (IGF-I). While the latter inhibits GH secretion by a dual effect on both the pituitary and the hypothalamus (8), no direct effect of GH on the somatotroph has been observed in in vitro studies (9, 10). The absence of a direct effect of GH on the pituitary indicates that the inhibition of GH secretion observed after GH administration without elevation of plasma IGF-I levels (11,12) is mediated by the central nervous system (CNS). Acetylcholine modulation of GH secretion in man has been extensively documented. Several studies have shown that cholinergic blockade inhibits spontaneous and exercise-, GH-releasing hormone (GHRH)-, insulin hypoglycemia-, arginine-, and clonidine-stimulated GH secretion (13-15). Conversely, enhanced cholinergic tone induced by acetylcholinesterase inhibitors augments spontaneous and GHRH- and clonidine-stimulated GH
Materials and Methods Subjects
Received August 17, 1990. Address all correspondence and requests for reprints to: Lawrence A. Frohman, M.D., Division of Endocrinology and Metabolism, University of Cincinnati College of Medicine, 231 Bethesda Avenue, ML 547, Cincinnati, Ohio 45267. * This work was supported in part by USPHS Grants DK-30667 and RR-00068 and a grant from Genentech, Inc.
Sixteen men, aged 18-34 yr (mean ± SE, 23.6 ± 1.1) participated in the study. Some of the subjects had participated in a previous research protocol (12) that included studies under conditions similar to those of studies in the present research protocol; data obtained at that time were included in the present study. All subjects were within 15% of ideal body weight and were receiving no medications. Each subject gave written informed consent to the protocols, which were approved by the Institutional Review Board. Protocols were performed in the University of Cincinnati General Clinical Research Center, with an interval of at least 7 days between studies. The order of the studies was randomly assigned. The subjects fasted from
1081
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 November 2015. at 13:32 For personal use only. No other uses without permission. . All rights reserved.
KELIJMAN AND FROHMAN
1082
midnight until completion of the study. Indwelling needles were inserted in both forearm veins between 0730-0830 h, and subjects remained recumbent and awake during the studies. Protocols Combined GHRH and TRH stimulation Six men received 6-h infusions of biosynthetic GH (Protropin, Genentech, Inc., South San Francisco, CA) at a rate (0.55 Mg/m2/min) we have previously shown to significantly inhibit plasma GH responses to GHRH stimulation (12) or 150 mmol/ L NaCl. Four hours after the initiation of the infusion, GHRH(1-44) (1 Mg/kg; Hoffmann-LaRoche, Inc., Nutley, NJ) and TRH (0.3 jug/kg; Thypinone, Abbott Laboratories, North Chicago, IL) were injected iv. A maximally stimulatory dose of GHRH was chosen to exclude the effects of endogenous GHRH secretion. These studies were repeated with the addition of 120 mg pyridostigmine administered orally 30 min before the GHRH/TRH injection. It has been previously shown that administration of this dose of pyridostigmine results in enhanced plasma GH responses (17). GH, GHRH, and TRH were diluted as previously described (12). Insulin hypoglycemia and arginine stimulation In similar protocols, regular insulin (0.1 U/kg, iv bolus injection; six subjects) or L-arginine hydrochloride (30 g infused iv over a 30-min period; six subjects) was substituted for GHRH plus TRH as the stimulus for hormone secretion. Blood glucose levels were measured at 15-min intervals after insulin injection.
JCE&M«1991 Vol 72 • No 5
Statistical analysis Plasma GH responses to GHRH, TRH, insulin, and arginine were quantified by measurement of peak incremental plasma hormone levels and integrated secretion, as determined by the trapezoidal method. Basal GH concentrations were defined as the mean of the plasma GH values during the 2-h period preceding the stimulus. Plasma GH, IGF-I, glucose, and FFA levels before stimuli administration in all protocols were pooled. Group differences were analyzed by a repeated measures twofactor analysis of variance (SAS Institute, Cary, NC) for multiple comparisons after logarithmic transformation when indicated. A one-tailed P value of less than 0.05 for effects of exogenous GH and pyridostigmine was considered significant.
Results Effects of exogenous GH and pyridostigmine on GH and TSH responses to GHRH and TRH Plasma GH responses and the integrated secretory response to GHRH plus TRH are shown in Fig. 1 and Table 1. Exogenous GH significantly inhibited GH secretory responses, expressed as both peak (41%; P = 0.01) and integrated (48%; P < 0.01) GH secretion. In contrast, pyridostigmine significantly enhanced both 40
GH (0.55 ug/m7min) or SAUNE PYRIDOSTIGMINE GHRH + TRH
30
Sample collection and processing Blood samples were collected in heparinized tubes every 15 min throughout each study for measurement of plasma GH, TSH, and PRL, as indicated below. Samples for glucose, IGFI, and FFA determinations were obtained immediately before the saline/GH infusion and immediately before GHRH plus TRH or insulin injection or arginine infusion. All blood samples were immediately chilled, centrifuged at 4 C, and stored at -20 C until assayed. Plasma hormone concentrations in each study were determined in a single assay. Plasma GH, TSH, and PRL were measured in duplicate by RIA, as previously described (19-21), with intra- and interassay coefficients of variation of 6.1% and 6.8%, 7.2% and 7.8%, 10.2% and 10.7%, and 4.5% and 9.1%, respectively. Plasma IGF-I was assayed by a modification of the method of Daughaday et al. (22); biosynthetic IGF-I (Amgen Biologicals, Thousand Oaks, CA) was used for iodination and as standard. AntiIGF-I serum was provided by the National Hormone and Pituitary Program, NIDDK (Baltimore, MD). The intra- and interassay coefficients of variation were 4.1% and 10.5%, respectively. Plasma glucose levels were measured by a glucose oxidase method. Serum FFA concentrations were determined by an enzymatic colorimetric method (Wako Pure Chemical Industries, Ltd., Osaka, Japan).
\
20
o.-o CONTROL (6)
10
a--* GH(7) •— PYR(6) ^
30
GH + PYR (6)
240
270
300 MINUTES
330
360
390
FIG. 1. Effects of exogenous GH and pyridostigmine (PYR), alone and in combination, on the plasma GH response (A) to GRH (1 fig/kg, iv). Shown are the mean ± SE. The number of subjects in each group is shown in parentheses. Baseline plasma GH levels were 0.4 ± 0.04 Mg/L (control), 3.6 ± 0.1 Mg/L (GH infusion), 0.4 ± 0.03 Mg/L (pyridostigmine), and 4.0 ± 0.4 ^g/L (GH infusion plus pyridostigmine). The increase in basal GH levels attributable to the infusion of exogenous GH was subtracted from the measured values to facilitate comparison of the responses to stimuli.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 November 2015. at 13:32 For personal use only. No other uses without permission. . All rights reserved.
CHOLINERGIC MODULATION OF GH FEEDBACK
1083
TABLE 1. Effect of exogenous GH and pyridostigmine on plasma GH responses
Stimulus GHRH (1 Mg/kg) No. of subjects Peak increment iixg/h) Integrated response (^g/L • min) Insulin (0.1 U/kg) No. of subjects Peak increment (ng/L) Integrated response (^g/Lmin) Arginine (30 g) No. of subjects Peak increment (^g/L) Integrated response (fig/L • min)
Combined effects0
Control
GH
Pyridostigmine
GH + pyridostigmine
6 21.6 ± 5.0 1434 ± 332
7 3.6 ± 0.1 353 ± 47
6 29.8 ± 7.2 2436 ± 580
6 25.5 ± 5.4 1761 ± 406
0.01
