Arginine but not pyridostigmine, a cholinesterase inhibitor, enhances the GHRH-induced GH rise in patients with anorexia nervosa

Abstract Pirenzepine, a muscarinic antagonist probably acting via stimulation of hypothalamic somatostatin release, abolishes the growth hormone releasing hormone (GHRH)-stimulated growth hormone (GH) rise in normal subjects but only blunts it in patients with anorexia nervosa (AN). This finding suggested the existence in AN of an alteration of cholinergic system and/or somatostatinergic tone. To further investigate these mechanisms, in 11 AN women patients (age 18.8 ± 0.9 years; BMI 13.4 ± 0.4 we studied the GH response alone (1 μg/Kg IV as a bolus at 0 min) and combined with pyridostigmine (PD, 120 mg orally, 60 min before GHRH administration), a cholinesterase inhibitor, or arginine (ARG 30 g infused over 30 min starting at 0 min), two compounds probably acting via inhibition of hypothalamic somatostatin (SS) release. The GH response to GHRH preceded by a previous (120 min before) neurohormone administration also was studied. All these tests also were performed in 20 normal age-matched women (age22.0 ± 1.8 yrs; BMI 20.1 ± 2.4). Basal serum GH levels were higher in AN patients than in normal volunteers (NV) (10.3 ± 3.4 versus 2.8 ± 0.3 μg / L ; p 0.001), whereas plasma IGF-1 levels were lower in AN patients than in NV (43.3 ± 10.6 versus 172.4 ± 13.9 μg / L ; p 0.00001). In AN patients, GHRH administration induced a GH rise higher, though not significantly, than that in NV [Δ area under the curve (AUC) 11173.6 ± 167.6 versus 834.6 ± 188.1 μg / L / h ). The GH response to the second of two consecutive GHRH boluses was lower (p 0.01) than that of the first one either in AN patients or in NV (67.6 ± 27.4 and 53.1 ± 25.7 μg / L / h , respectively). PD administration failed to modify the GHRH-induced GH rise in AN patients (1370.4 ± 228.4 versus 1268.8 ± 205.8 μg / L / h ), but enhanced it in NV 2285.4 ±234.6 versus 1070.1 ± 159.4 μ / L / h ; p 0.01). The GH response to PD + GHRH in AN patients was lower (p 0.01) than that in NV, ARG administration enhanced the GHRH-induced GH rise either in AN patients (3843.0 ± 659.7 μg / L / h ; p 0.02 versus GHRH alone) or in NV (3346.6 ± 424.1 μ / L / h ; p 0.01 versus GHRH alone), the two responses being similar. Our results demonstrate that, although ARG fails to modify the GHRH-induced GH response in NV, in AN the GH response to GHRH is inhibited by a previous neurohormone administration. These findings indicate the existence in AN of somatotroph refractiries to cholinergic influence but not to metabolic fuels such as arginine, whereas the negative GH-autofeed-back mechanism is preserved. Somewhat specific alterations of the somatostatin-mediated cholinergic influence or, alternatively, different actions for PD and ARG may be hypothesized.