J. Endocrinol. Invest. 13: 653-656, 1990

Interaction of glucose and pyridostigmine on the secretion of growth hormone (GH) induced by GH-releasing hormone (GHRH) G. Oelitala, P. A. Tomasi, M. Palermo, and P. Fresu Istituto di Ematologia e di Endocrinologia, Cattedra di Endocrinologia, Universita di Sassari, Viale S. Pietro 12, 07100 Sassari, Italy.

ABSTRACT. In order to investigate the mechanisms by which hyperglycaemia induces an inhibition of GHRH-induced GH release, we gave the following treatments to seven normal men: a) GHRH 100 ILg iv; b) pyridostigmine (PD) 120 mg po 60 min before GHRH; c) glucose 250 mgt kg iv as a bolus (10 min before GHRH) plus 10 mgtkgtmin until the end of the test; d) glucose, pyridostigmine and GHRH as above. Glucose sig-

nificantly reduced GHRH-stimulated GH levels, whereas PD Significantly enhanced them. When PD and glucose were given together, the effect on GHRHstimulated GH secretion was not different from the algebraic sum of the single effects of the two substances. Thus glucose seems to be able to exert its inhibition, at least partially, also when pyridostigmine is coadministered.

INTRODUCTION Variations of glycemia exert marked effects on GH release both in experimental animals and in humans (1). In particular, hypoglycemia is a potent GH releaser and hyperglycemia blunts basal and stimulated GH concentrations in man (1, 2). It is apparent that rapid variations of glycemia, rather than absolute glucose concentrations, constitute the stimulus for GH secretion or inhibition (3, 4), as is also shown by the normal response of GH to GHRH in diabetic subjects (5). The site at which glycemic variations exert their effects is not yet completely elucidated. Studies in vitro do not support the hypothesis of a direct effect on the pituitary (6), and experiments in animals suggest that the effects of glycemic variations may be mediated by somatostatin neurons (7, 8); this is also supported by the observation that hyperglycemia can at least partially inhibit the GH release induced by maximal doses of GHRH in man (9). In order to elucidate whether there was an interaction between the glucose-induced inhibition of GH

release after GHRH and the hypothalamic cholinergic system, we studied the effect of pyridostigmine, an acetylcholine-esterase inhibitor, on glucose- induced suppression of GHRH-stimulated GH release in normal men. It is known that cholinergic antagonists may block the effect of many GH-releasing agents, while agonists actually increase basal or GHRH-stimulated GH secretion (10). Cholinergic neurons are suggested to directly modulate the activity of the hypothalamic somatostatinergic cells. MATERIALS AND METHODS Seven healthy male subjects, aged 18-33 yr, volunteered for this study. They were all within ± 10% of their ideal body weight, and did not take drugs in the previous week. The subjects were placed in bed at 08:30 h, after an overnight fast, and were not allowed to stand, eat, smoke or drink anything except water for the duration of the test. An iv needle was placed in a forearm vein for blood sampling, and another iv cannula was inserted in the contralateral arm for glucose (or 150 mmol NaCI) administration. All subjects were then given, in different occasions (with an interval of at least 7 days) and in randomized order, the following treatments: 1) GHRH (1-29) NH2 (GHRH, KabiVitrum, Stockholm, Sweden) 100 ILg in 2 ml sterile water, given iv as a bolus;

Key-words: GH, GHRH, glucose, pyridostigmine, acetylchOline. human. hyperglycaemia. Correspondence: Prof. G. Delitala. Cattedra di Endocrinologia. Viale S. Pietro 12. 07100 Sassari. Italy.

Received December 20. 1989; accepted June 18. 1990

653

G. Delitala, P.A. Tomasi, M. Palermo, et al.

2) GHRH as above, plus pyridostigmine (PD, Mestinon, Roche) 120 mg given po 60 min before GHRH; 3) GHRH as above, plus glucose 250 mglkg body weight given iv as a bolus of 33% glucose in water, 10 min before GHRH, and immediately followed by an iv continuous infusion of 10 mglkglmin until the end of the test. This dose was calculated according to DeFronzo et al. (11 ). 4) GHRH as above, plus PD as above, plus glucose as above. Blood samples were drawn in EDTA tubes basally, immediately before GHRH, and at 15, 30, 45 and 60 min after GHRH administration. Glucose was assayed (for tests 3 and 4) by an automated system using glucose-oxidase. After centrifugation, plasma was immediately stored at -20 e. Plasma GH was assayed with a commercial RIA kit (Biomerieux, eharbonnieres-Ies-Bains, France). All samples coming from the same subject were assayed within the same batCh and with a single standard curve; the kit's intraassay coefficient of variation was 5%. Plasma GH concentrations are expressed in I-'g/l (where 1 ~g/l=1 ng/ml); results are shown as means ± SE. Integrated concentrations (Ie) of GH were determined according to the polygonal method. Since GH values do not show a uniform variance, particularly after stimulation, the appropriate logarythmic transformation was applied on the data before statistical analysis; however original data are shown as results. Statistical analysis for GH values was performed by means of two-way factorial analysis of variance (factors: glucose and PD); this technique allows for a determination of interaction significance, i.e. gives us a precise indication about the presence or absence of interference between the two factors. Statistical analysis for blood glucose values in tests 3 and 4 was carried out with paired Student's t tests. RESULTS The subjects did not experience any side-effect during the tests. Plasma GH concentrations after the four treatments are shown in Figures 1 and 2, and glycemia values for treatments 3 and 4 are shown in Figure 3. stood glucose levels were not different in treatments 3 and 4, both on integrated concentration (212 ± 19 vs 206 ± 18 mg I dl) and at all times (p > 0.05

60

....J

40

"C'I

~ :J: t.:l

20 PO

1

0 -60

o

15

30

45

60 min.

Fig. 1 - Plasma GH concentrations (means ± SE) following GHRH (100 pg), GHRH plus pyridostigmine (120 mg po 60 min before GHRH), GHRH plus glucose (250 mg/kg iv 10 min before GHRH, followed by 10 mg/kg/min for 70 min), GHRH plus pyridostigmine and glucose.

by Student's paired t test}. Mean plasma GH levels at peak (P) and mean plasma GH Ie after GHRH alone were respectively 25.75 ± 5.1 I-'g/l and 20.06 ± 3.4. Glucose administration Significantly reduced these levels: after glucose + GHRH, plasma GH levels were 10.79 ± 2.6 I-'g/l at peak, and 9.09 ± 1.3 on Ie (glucose main effect for les and peaks: p < 0.05). Pyridostigmine greatly increased GH levels, with a

lnIatn I8o'1ts (G+-RH giYan

n aJ trt)

Fig. 2.- Plasma GH integrated concentrations following GHRH alone, GHRH plus pyridostigmine, GHRH plus glucose, and GHRH plus pyridostigmine and glucose.

654

Pyridostigmine, glucose and GH release

mean peak concentration reaching 54.26 ± 3.2 and an Ie of 44.33 ± 4.0 ~g/I (PO main effect for les and peaks: p < 0.01). When the two substances were given together (in association with GHRH), GH levels rose to 44.87 ± 8.8 ~g/I at peak, with an Ie of 38.65 ± 6.8. These levels were close, but inferior to those obtained after PO alone (54.26 ± 3.2 ~g/I at peak, 44.33 ± 4.0 Ie). Statistical analysis on both les and peak values disclosed a nonsignificant interaction between the factors. This means that the effect when both factors are given is very similar to the algebraic sum of the two main effects (in our case, a negative effect is present: glucose), i.e. the two factors do not seem to interfere with each other and thus they appear to have independent mechanisms of action.

lamic somatostatinergic cells has been demonstrated in vitro (7,8). Pyridostigmine, a compound which does not readily cross the blood-brain barrier, stimulates muscarinic receptor activity by inhibiting acetylcholine-esterase. Previously published work also has demonstrated that PO counteracts the inhibitory action of exogenous GH on GHRH-stimulated GH secretion (12), and the reduced responsiveness of GH release following repeated GHRH injections (13). Since both situations are likely mediated by a feedback blockade exerted by somatostatin on GH secretion, the data, taken together, suggest that pyridostigmine may inhibit somatostatin release from the hypothalamus. Our findings show that pyridostigmine, a cholinergic agonist, is able to counteract, albeit not completely, the effect of glucose on GHRH-induced GH secretion. Since PO is believed to act via somatostatinergic neurons (12, 13), this would suggest somatostatin as the mediator of the glucose effect. However the incomplete reversal of the effect of glucose, shown by the no'nsignificant interaction of the two factors, suggests that additional mechanisms might be involved in the glucose-induced inhibition of the GHRH-stimulated GH secretion. Before submission of this paper, a similar study was published by other authors (14, 15). However, they used orally administered glucose for suppression of GH secretion induced by GHRH, and blood glucose levels started to fall at the time of the injection of GHRH. In the experiment reported by Balzano et al. (15) pyridostigmine was only administered with glucose (and GHRH) but not alone, which precludes a more sensitive analysis of the data; furthermore, the dose of GHRH is not stated in the paper. On the contrary, our results are in agreement with the data shown by Penalva et al. (14), but the different statistical approach does not allow us to reach a similar interpretation of the data. In conclusion, PO administration appears to potentiate the GH release induced by GHRH when given with glucose, but its effect is of smaller importance than when PO is given alone. According to our statistical approach we suggest that glucose-induced inhibition of GHRH-stimulated GH release likely involves the participation of somatostatinergic and cholinergic neurons. However other mechanisms may possibly participate in this homeostatic process.

~g/I,

300' 250 ....J "0

200 'm E

0/ 150 en a u :J 100 ......

t:n

50

o -10

0

15

30

45

60

min.

Fig. 3 - Blood glucose concentrations (means ± SE) during glucose infusion in the GHRH alone and the GHRH plus pyridostigmine experiments.

DISCUSSION Our results confirm that glucose is able to inhibit the GHRH-induced GH secretion in man. They also confirm that, on the other hand, pyridostigmine further increases the effect of GHRH on GH secretion. Glucose does not seem to act directly on pituitary cells, as shown by in vitro studies (6). Thus it likely has an action on either GHRH or somatostatin. Since glucose is able to blunt the GH release induced by a maximal dose of GHRH, the latter hypothesis looks more probable. In keeping with this hypothesis, a direct effect of glucose on hypotha-

655

G. Delitala, P.A. Tomasi, M. Palermo, et al.

ACKNOWLEDGMENTS We are grateful to KabiVitrum (Stockholm, Sweden) and to Pierrel (Milan. Italy) for the kind supply of GHRH(1-29)NH 2 . This work was supported by CNR grant 89025804 and a grant from Ministero della Pubblica Istruzione to G.D.

incubated rat hypothalamus and cerebral cortex. Effect of glucose and glucoregolatory hormones. J. Clin. Invest. 69: 1293, 1982. 9. Masuda A., Shibasaki T., Nakahara M., Imaki T., Kiyosawa Y., Jibiki K., Demura H., Shizume K., Ling M.

REFERENCES 1. Daughaday W.H. The anterior pituitary. In: Wilson J.D., Foster DW. (Eds.), Williams' Textbook of Endocrinology. W.B. Saunders, Philadelphia, 1985, p. 580. 2. Roth J., Glick S.M., Yalow R.S., Berson SA Hypoglycaemia: a potent stimulus to secretion of growth hormone. Science 140: 987, 1963. 3. Hunter W.M., Willoughby J.M.T., Strong J.A. Plasma insulin and growth hormone during 22-hour fasts and after graded glucose loads in six healthy adults. J. Endocrinol. 40: 297, 1968. 4. Hunter W.M. Growth Hormone. In: loraine J.A., Bell E.T. (Eds.), Hormone assays and their clinical applications. Churchill Livingstone, Edinburgh, 1976, p. 223. 5. Press M., Tamborlane WV, Thorner M.D. Pituitary response to growth hormone-releasing factor in diabetes. Failure of glucose-mediated suppression. Diabetes 33: 804, 1984. 6. Quabbe H.J. The endocrinology of growth hormone-secreting microadenomas. In: Faglia G., Giovannelli M.A., Macleod R.M. (Eds.), Pituitary microadenomas. Academic Press, New York, 1980, p. 229. 7. lengyel A.N.J., Grossman A., Nieuwenhuyzen-Kruseman A.C., Ackland J., Rees L., Besser G.M. Glucose modulation of somatostatin and lH-RH release from rat hypothalamic fragments in vitro. Neuroendocrinology 39: 31, 1984.

The effect of glucose on growth hormone-releasing hormone-mediated GH secretion in man. J. Clin. Endocrinol Metab. 60: 523, 1985. 10. Delitala G., Tomasi P., Virdis R. Neuroendocrine regulation of human growth hormone secretion. Diagnostic and clinical applications. J. Endocrinol. Invest. 11: 441, 1988. 11. DeFronzo RA, Tobin J.D., Andres R. Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am. J. Physiol. 237: E214, 1979. 12. Ross R.J.M., Tsagarakis S., Grossman A., Nhagafoong l., Touzel R.J., Rees L.H., Besser G.M. GH feedback occurs through modulation of hypothalamic somatostatin under cholinergic control: studies with pyridostigmine and GHRH. Clin. Endocrinol. (Oxf.) 27: 727, 1987. 13. Massara F., Ghigo E., Molinatti P., Mazza E., locatelli V., Muller E.E., Camanni F. Potentiation of cholinergic tone by pyridostigmine bromide re-instates and potentiates the growth hormone responsiveness to intermittent administration of growth hormone-releasing factor in man. Acta Endocrinol. (Copenh.) 113: 12, 1986. 14. Penalva A., Burguera B., Casabiell X., Tresguerres JAF., Dieguez C., Casanueva F.F. Activation of cholinergic neurotransmission by pyridostigmine reverses the inhibitory effect of hyperglycaemia on GHRH-induced GH secretion in man. Does acute hyperglycaemia act through hypothalamic release of somatostatin? Neuroendocrinology 49: 551, 1989. 15. Balzano S., loche S., Murtas M.L., Pintor C., Martino E. Potentiation of cholinergic tone counteracts the suppressive effect of oral glucose administration on the GH response to GHRH in man. Horm. Melab. Res. 21: 52, 1989.

8. Berelowitz M., Dudlak D., Frohman L.A. Release of somatostatin-like immunoreactivity from

656