Psychoneuroendocrinology 25 (2000) 179–185

Muscarinic cholinergic mediation of the GHresponse to gamma-hydroxybutyric acid:neuroendocrine evidence in normal and

parkinsonian subjects

Riccardo Volpi a,*, Paolo Chiodera a, Paolo Caffarra b,Augusto Scaglioni c, Laura Malvezzi c, Antonio Saginario d,

Vittorio Coiro a

a Department of Internal Medicine and Biomedical Sciences, School of Medicine, Uni6ersity of Parma,Via A. Gramsci 14, I-43100 Parma, Italy

b Institute of Neurology, School of Medicine, Uni6ersity of Parma, Via del Quartiere 4,I-43100 Parma, Italy

c Di6ision of Neurology, Hospital of Fidenza, Via Borghesi 1, Fidenza, Italyd Di6ision of Psychiatry, Hospital of Piacenza, Piacenza, Italy

Received 23 March 1999; accepted 29 June 1999

Abstract

We have recently reported that parkinsonian patients show a significant GH response togamma-hydroxybutyric acid (GHB), but not to gamma-aminobutyric acid (GABA)-ergic drugadministration. In order to establish whether muscarinic cholinergic receptors mediate the GHsecretion induced by GHB, normal men and parkinsonian patients were tested with GHB bothin the absence and in the presence of the anticholinergic agent, pirenzepine. Both normalcontrols and parkinsonian patients showed a significant serum GH rise in response to GHB(25 mg/kg body weight p.o.) even though a slightly, but significantly lower response wasobserved in parkinsonian patients. Pretreatment with pirenzepine (100 mg p.o. 2 h before GHB)completely suppressed the GHB-induced GH release in both normal controls and parkinsonianpatients. These data indicate that a cholinergic mechanism mediates the GH response to GHBin normal men. In addition the data indicate that this pathway is preserved in the parkinsonianbrain. © 2000 Elsevier Science Ltd. All rights reserved.

Keywords: Parkinson’s disease; GH; Gamma-hydroxybutyric acid

www.elsevier.com/locate/psyneuen

* Corresponding author. Tel.: +39-0521-292461; fax: +39-0521-290786.E-mail address: volpi@ipruniv.cce.unipr.it (R. Volpi)

0306-4530/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.PII: S0306 -4530 (99 )00048 -7

180 R. Volpi et al. / Psychoneuroendocrinology 25 (2000) 179–185

1. Introduction

Both gamma-aminobutyric acid (GABA) and its breakdown product gamma-hy-droxybutyric acid (GHB) are well-known neurotransmitters and neuromodulatorsin the central nervous system. However, to have structural similarities and closebiochemical relationships does not necessarily mean that the two substances shareactivities in the control of the same neuroendocrine pathways. In fact, either similaror different neuropharmacological and neurophysiological effects have been re-ported after GHB and GABAergic drugs administration (for review see Snead,1978). Therefore, the mechanism of action of GHB may differ from that of GABA.This is evident, when GHB and GABA actions are different, but it is not so whenthe two substances produce similar effects.

GHB and GABA show similar significant stimulatory effects on GH secretion innormal human subjects. Surprisingly, we have recently observed that in parkinso-nian patients the GH releasing action of GABA was abolished, whereas GHB-in-duced GH release was preserved. These findings suggested that GHB and GABAstimulate GH through different pathways, that are differently affected by Parkin-son’s disease. We supposed that GHB stimulates GH secretion through a choliner-gic pathway, because GHB is known to modify cholinergic neurotransmission inthe brain (Volpi et al., 1997) and the cholinergic system plays an important role inthe control of GH secretion (Richardson et al., 1980).

In order to verify this hypothesis, in the present study we tested the GH-releasingaction of GHB in normal and parkinsonian patients in basal conditions and afterblockade of muscarinic cholinergic receptors with pirenzepine.

2. Materials and methods

Ten men (53–70 years old; mean9SE: 6391.6) affected by de novo Parkinson’sdisease were randomly chosen to participate in this study. The Ethical Committeefor clinical trials of the Department of Internal Medicine and Biomedical Sciencesof the University of Parma reviewed the protocol of the study and the subjects gavetheir informed consent. The diagnosis of Parkinson’s disease had been establishedby the presence of clinical features (resting tremor, bradykinesia and rigidity). Themean (9SE) duration of the disease was 22.695.1 months. The severity of themedical stage, assessed with the Hoen–Yahr scale (Hoen and Yahr, 1967), was1.990.2. Patients with secondary parkinsonism or with clinical evidence of othermajor neuroanatomical system involvements were excluded. All patients wereassessed with the hospital anxiety and depression scale (HADS) (Zigmond &Snaith, 1983), the research diagnostic criteria (RDC) (Spitzer et al., 1978), and withthe Hamilton depression rating scale (HDRS) (Hamilton, 1960). A psychiatrist(S.A.) was specifically trained to administer RDC criteria, the HRDS and theHADS. The training consists in the administration of the RDC and the depressionscales in five patients (affected by major depression or disthymia). The interviews

181R. Volpi et al. / Psychoneuroendocrinology 25 (2000) 179–185

were audiotaped and reviewed by two senior psychiatrists. A satisfactory reliabilitywas obtained during the training for the RDC criteria (k=0.76), HDRS score(k=0.82) and the HADS score (k=0.80). No depressive disorder according to theRDC criteria or depressive symptomatology was found as shown by a score ofHDRS lower than ten and a score of the HADS depression subscale lower thanseven in all patients. None of the subjects resulted to be affected by depression. Allof them were within 10% of their ideal body weight. They were fully ambulatory,well nourished and without clinical or laboratory evidence of renal, neoplastic orendocrine–metabolic diseases.

Ten normal weight male volunteers aged 51–72 years (mean9SE: 64.792.1)were tested as controls.

All subjects were tested with g-hydroxybutyric acid (GHB), GHB plus piren-zepine and placebo. Tests were carried out in random order at weekly intervals.

3. Experimental procedure

At 0900 h of the experimental day, a 19-gauge intravenous indwelling needle wasinserted into an antecubital vein of subjects lying in the recumbent position andfasting from the previous evening. The needle was kept patent with a slow infusionof normal saline. A basal blood specimen was taken 15 min after the insertion ofthe cannula (time 0).

3.1. GHB test

GHB (Alcover, Laboratorio Farmaceutico CT, Sanremo, Italy) at a dose of 25mg/kg body weight or a placebo was administered p.o. at time 0 just after bloodsampling. Further blood samples were taken at 15, 30, 45, 60, 90 and 120 min afterGHB administration.

3.2. GHB plus pirenzepine test

This test was performed as previously described for the GHB test except for theoral administration of pirenzepine (Gastrozepin, Boehringer Ingelheim, Firenze,Italy) (100 mg p.o.) 2 h before GHB test. In the GHB test a placebo was giveninstead of pirenzepine.

3.3. Control test

Placebos were administered two hours earlier and at time 0 after blood sampling.Further samples were taken at time 15, 30, 45, 60, 90 and 120 min after the secondplacebo administration.

Sampling time points were established according to previous studies of the effectsof GHB (Gerra et al., 1994; Volpi et al., 1997) on GH secretion.

Blood pressure and heart rate were monitored at each sampling time.

182 R. Volpi et al. / Psychoneuroendocrinology 25 (2000) 179–185

4. Assays

Blood samples were collected and centrifuged cold; serum was stored at −20°Cuntil assayed. Serum GH concentrations were measured with a specific RIA, usingmaterials supplied by Ares Serono (Milan, Italy). All samples were analyzed in thesame assay and in duplicate. Sensitivity, intra-assay and inter-assay coefficients ofvariation were: 0.5 ng/ml, 2.9% and 6.5%, respectively.

Statistical analysis was performed with Mann–Whitney U test and analysis ofvariance (ANOVA) or analysis of covariance (ANCOVA), as appropriate. AN-COVA was performed to exclude the possible interference of variations in severityof the medical stage of Parkinson’s disease in the analysis and interpretation of thevarious tests.

The area under the curve (AUC) was calculated by trapezoid method.Data are reported as the mean9SE.

5. Results

The administration of placebos alone was unable to change the basal secretion ofGH in normal controls (Fig. 1) and in parkinsonian subjects (Fig. 2). In contrast,when GHB was given alone, significant increments in serum GH levels wereobserved in both groups (normal controls: mean peak levels at 60 min, PB0.01(Fig. 1); parkinsonian patients: mean peak levels at 45 min, PB0.01 (Fig. 2)).However, when the areas under the curve of the GH response to GHB in normaland parkinsonian patients were compared, a significantly lower GH response wasobserved in parkinsonian subjects (PB0.05 vs. normal controls) (Fig. 3).

Fig. 1. Effect of gamma-hydroxy-butyric acid (GHB), GHB plus pirenzepine and placebo on serum GHlevels in normal controls. Each point represents the mean9SE of 10 observations.

183R. Volpi et al. / Psychoneuroendocrinology 25 (2000) 179–185

Fig. 2. Effect of gamma-hydroxy-butyric acid (GHB), GHB plus pirenzepine and placebo on serum GHlevels in parkinsonian patients. Each point represents the mean9SE of ten observations.

The concomitant administration of pirenzepine completely abolished the GHB-induced GH rise in both normal controls (GHB plus pirenzepine test vs. GHB testF=22.71, PB0.001) (Fig. 1) and parkinsonian subjects (GHB plus pirenzepine testvs. GHB test F=20.60, PB0.001) (Fig. 2).

5.1. Side effects

No side effects were observed in any subjects after GHB and pirenzepineadministration.

Fig. 3. Serum GH responses (area under the curve (A.U.C.): mean9SE) in ten normal controls (N.C.)and in ten parkinsonian patients (PARK).

184 R. Volpi et al. / Psychoneuroendocrinology 25 (2000) 179–185

6. Discussion

The data reported here confirm our previous finding of a significant GH responseto GHB in Parkinson’s disease. As the normal subjects, the parkinsonian patientsshowed a significant GHB-induced GH increment, however, the GH secretorypattern was different in the two groups, because the normal subjects showed aquicker significant rise (at 15 min) than parkinsonian patients (at 30 min) and themean peak response was reached at 60 min in the normal controls and at 45 minin the parkinsonian subjects. Furthermore, measurements of the areas under thecurve showed that the GH response to GHB was lower in parkinsonian than innormal subjects.

When the normal subjects were treated with pirenzepine before GHB administra-tion, no significant rise in GH secretion was observed, suggesting that a cholinergicpathway mediates GHB action. In contrast, we have previously reported that insimilar experimental conditions pirenzepine is unable to change the GH releasingeffect of the GABAergic B receptor agonist baclofen, arguing against a cholinergicrole in the control of the GH response to GABA (Coiro et al., 1985). Theadditional results of this study show that also in Parkinson’s disease the muscariniccholinergic mechanism underlying GHB activity is well preserved and can beinhibited by pirenzepine.

It is unlikely that the abovementioned differences observed in the GH responseto GHB between normal and parkinsonian subjects were due to cholinergicreceptor changes, because muscarinic cholinergic receptor density and binding inautopsied samples from various cerebral nuclei and structures of parkinsonianpatients did not show significant decrements versus the normal condition (Nishinoet al., 1988; Rinne et al., 1989). The different pattern between the two groups mightbe due to alterations in acethylcholine content and/or disposal at hypothalamic-pi-tuitary level in parkinsonian patients.

A direct effect of GABA and GABAergic drugs at the pituitary level has beenexcluded by ‘in vitro’ studies (Flock et al., 1984; Racagni et al., 1982; Vijayan andMcCann, 1978). In contrast, it is unknown whether GHB stimulates GH secretiondirectly from the pituitary gland. In the light of our present results, this possibilitycannot be excluded because cholinergic receptors have been found in anteriorpituitary cell membranes (Casanueva et al., 1986) and acethylcholine has beenreported to cause a dose-related stimulation of GH in dispersed pituitary cells(Richardson et al., 1980). Furthermore, pirenzepine does not cross the blood brainbarrier (Hammer & Koss, 1979) and it is known to abolish pituitary GH secretioninduced by GH–RH (Massara et al., 1984). Therefore, it is possible that not onlyGHB stimulation of GH secretion, but also pirenzepine inhibitory effect take placeat the pituitary level.

GABA likely stimulates GH secretion by acting at hypothalamic level, where itis thought to inhibit somatostatin release by reducing dopaminergic action (for rev.see Racagni et al., 1982). Also GHB might be hypothesized to stimulate GHsecretion at hypothalamic level, suppositively through cholinergic mediation. Infact, it has been reported that acetylcholine inhibits the release of somatostatin

185R. Volpi et al. / Psychoneuroendocrinology 25 (2000) 179–185

from the hypothalamus in vitro through a muscarinic mechanism (Richardson etal., 1980). This action might take place at hypothalamic sites located outside theblood brain barrier, such as the median eminence, where pirenzepine anticholinergicaction may be supposed to enhance endogenous somatostatin release and thusinhibit the GH response to GHB.

In the light of the abovementioned hypotheses on the mechanism(s) of action ofGHB on GH secretion, our results might indicate that the cholinergic mechanismcontrolling GH secretion at pituitary level and/or somatostatin release at hypotha-lamic level is (are) preserved in parkinsonian patients.

References

Casanueva, F.F., Villanueva, L., Dieguez, C., Cabranes, J.A., Diaz, Y., Szoke, B., Scanlon, M.F.,Schally, A.V., 1986. Atropine blockade of growth hormone (GH)-releasing hormone-induced GHsecretion in man is not exerted at pituitary level. J. Clin. Endocrinol. Metab. 62, 186–191.

Coiro, V, Volpi, R, Dasso, L, Marehesi, C, Castelli, A, Rossi, G, Camellini, L, Chiodera, P, 1985.Cholinergic-muscarinic receptors do not mediate growth hormone release in response to baclofen, ag-aminobutyric acid agonist, in humans. Horm. Metabol. Res. 17, 260–261.

Flock, J., Acs, Z., Makara, G.B., Erdo, S.L., 1984. Site of a-aminobutyric acid (GABA)-mediatedinhibition of growth hormone secretion in the rat. Neuroendocrinology 39, 510–516.

Gerra, G., Marcato, A., Fertonani Affini, G., Avanzini, P., Maestri, D., Fontanesi, B., Caccavari, R.,Delsignore, R., 1994. Gamma hydroxybutiric acid (GHB) and neuroendocrine function in humans.Neuroendocrinol. Lett. 7, 55–63.

Hamilton, M., 1960. A rating scale for depression. J. Neurol. Neurosurg. Psychiatry 23, 56–62.Hammer, R., Koss, F.W., 1979. The pharmacokinetic profile of pirenzepine. Scand. J. Gastroenterol. 14

(Suppl. 57), 1–6.Hoen, M.M., Yahr, M.D., 1967. Parkinsonism: onset progression and mortality. Neurology 17,

427–441.Massara, F., Ghigo, E., Goffi, S., Molinatti, G.M., Muller, E.E., Camanni, F., 1984. Blockade of

hp-GRF-40-induced GH release in normal men by a cholinergic muscarinic antagonist. J. Clin.Endocrinol. Metab. 59, 1025–1026.

Nishino, N., Fujiwara, H., Noguchi-Kuno, S.A., Tanaka, C., 1988. GABA A receptor but notmuscarinic receptor density was decreased in the brain of patients with Parkinson’s disease. Jpn. J.Pharmacol. 48, 331–339.

Racagni, G., Apud, J.A., Cocchi, D., Locatelli, V., Muller, E.E., 1982. GABAergic control of anteriorpituitary hormone secretion. Life Sci. 31, 823–838.

Richardson, S.B., Hollander, C.S., D’Eletto, R., Greenleaf, P.W., Thaw, C., 1980. Acetylcholine inhibitsthe release of somatostatin from the rat hypothalamus in vitro. Endocrinology 107, 122–129.

Rinne, J.O., Lonnberg, P., Marjamak, P., Rinne, U.K., 1989. Brain muscarinic receptor subtypes aredifferently affected in Alzheimer’s disease and Parkinson’s disease. Brain Res. 483, 402–406.

Snead, O.C., 1978. Gamma hydroxybutyrate in the monkey. Neurology 28, 636–642.Spitzer, R.L., Endicott, J., Robins, E., 1978. Research diagnostic criteria: rationale and reliability. Arch.

Gen. Psychiatry 35, 773–782.Vijayan, E.E., McCann, S.M., 1978. Effects of intraventricular injection of gamma-aminobutyric acid

(GABA) on plasma growth hormone and thyrotropin in conscious ovariectomized rats. Endocrinol-ogy 103, 79–83.

Volpi, R., Chiodera, P., Caffarra, P., Scaglioni, A., Saccani, A., Coiro, V., 1997. Different controlmechanisms of growth hormone (GH secretion between a-amino- and a-hydroxy-butyric acid:neuroendocrine evidence in Parkinson’s disease. Psychoneuroendocrinology 22, 531–538.

Zigmond, A.S., Snaith, R.P., 1983. The hospital anxiety and depression scale. Acta Psychiatr. Scand. 67,361–370.