GUALILLO, ORESTE, JORGE E. CAMINOS, RUBÉ N NOGUEIRAS, LUISA M. SEOANE, EVA ARVAT, EZIO GHIGO, FELIPE F. CASANUEVA, AND CARLOS DIÉ GUEZ. Effect of food restriction on ghrelin in normalcycling female rats and in pregnancy. Obes Res. 2002;10: 682-687. Objective: Ghrelin is a 28-amino-acid acylated peptide that was recently identified as the endogenous ligand for the growth hormone secretagogue receptor. Previous studies have shown that ghrelin potently increases growth hormone release and food intake. The aim of this study was to clarify the physiological implications of ghrelin in the regulation of energy balance, by assessing the effect of undernutrition throughout 21 days in normal-cycling and pregnant rats on ghrelin. Research Methods and Procedures:We have determined ghrelin levels by radioimmunoassay and gastric ghrelin mRNA expression by Northern blot analysis during 21 days of chronic food restriction (30% of ad libitum available diet) in normal-cycling female rats and in pregnancy. Results: Our results show that chronic food restriction led to an increase in plasmatic ghrelin levels in normal-cycling female rats. In pregnancy, ghrelin plasmatic levels were enhanced particularly during the latter part of gestation (19 and 21 days) compared with pregnant rats with free access to food. Gastric ghrelin mRNA expression showed a similar expression pattern, being higher in the food-restricted group than in the group fed ad libitum, in normal-cycling as well as in pregnant rats.Discussion: These observations indicate that ghrelin plasmatic levels and ghrelin gastric mRNA are up-modulated during undernutrition in normal-cycling rats and in pregnancy. These findings suggest that increased ghrelin levels may have a role in mediating the physiological responses to undernutrition and could represent an adaptative response to prevent long-lasting alterations in energy balance and body weight homeostasis.

Objective: The recently isolated endogenous GH secretagogue, named ghrelin, is a gastric peptide of 28 amino acids with an n-octanoylation in the serine 3 that confers the biological activity to this factor. Ghrelin has been shown to directly stimulate GH release in vivo and in vitro and to be involved in the regulation of gastric acid secretion and motility. In the present work we have studied gender and gonadal dependency of ghrelin mRNA expression in rat stomach. Design and Methods: We analysed ghrelin mRNA expression in rat stomach by Northern blot analysis. We also examined the effect of gonadal steroid deprivation on ghrelin mRNA expression. Results and Conclusions:The results obtained showed clearly that ghrelin gastric mRNA expression increased with age in young rats (up to 90 days old) but exhibited no significant sex difference at each age tested. Ghrelin mRNA levels were lowest at postnatal day 9, reaching a stable level of expression at day 40 in both female and male rats, although the increase in female rats appears much more gradual than that in males. Moreover, neither ovariectomy nor orchidectomy significantly modified ghrelin mRNA gastric levels in adult rats. In conclusion, these data indicate that ghrelin mRNA expression is associated with age and that a progressive increase is present from the perinatal period up to a stable level after puberty. Gonadal hormones did not alter ghrelin mRNA levels. Taken together, these data showed that ghrelin mRNA levels in young rats are age but not gender dependent, and are not influenced by gonadal steroids.

There is evidence that GH secretion is reduced in normal elderly subjects as well as in patients with Alzheimer''s disease (AD). To clarify the mechanisms underlying this GH hyposecretory state in 14 elderly subjects (age 65–75 years) and 15 AD patients (age 61–78 years), we studied the effects of both pyridostigmine (PD, 120 mg orally), a cholinesterase inhibitor, and arginine (ARG, 0.5 g/kg i.v.), two substances likely acting via inhibition of hypothalamic somatostatin, on GH response to GHRH (1 (µg/kg i.v.). The GH response to PD alone was also studied. Twenty-two young healthy volunteers were studied as control group. Basal GH levels were similar in young, elderly and AD subjects (0.7 ± 0.2, 0.8 ± 0.2 and 0.9 ± 0.2 µg/1). IGF-I levels were lower (p < 0.005) in elderly (73.9 ± 8.2 µg/1) and in AD subjects (108.0 ± 5.9 µg/l) than in young subjects (288.7 ± 22.1 µg/l); however, they were higher (p < 0.01) in AD patients than in the elderly subjects. The PD-induced GH release did not significantly differ in young, elderly and AD subjects while the GH responses to GHRH in the elderly (AUC: 297.9 ± 49.2 µg/l/h) and in AD subjects (437.6 ± 93.5 µg/l/h) were lower (p < 0.01) than in young subjects (658.6 ± 100.1 µg/l/h). PD potentiated the GH response to GHRH both in elderly and in AD subjects (901.7 ± 222.4 and 1,070.3 ± 207.2 µg/l/h, p < 0.005) but these responses were lower (p < 0.0001) than those recorded in young subjects (2,041.1 ± 245.6 µg/l/h). ARG potentiated the GHRH-induced GH rise both in elderly and in AD subjects (1,545.2 ± 246.0 and 1,659.3 ± 196.8 µg/l/h,p < 0.001) but in this case, the GH response to GHRH + ARG overlapped with that in young subjects (2,140.2 ± 229.5 µg/l/h). In contrast to young subjects, in elderly and in AD subjects, the potentiating effect of ARG on GHRH-induced GH rise was higher (p < 0.01) than that of PD. These results show that testing neural controls of GH secretion with different neuroactive substances does not allow to differentiate normal aging from AD. In both groups, somatotroph responsiveness to GHRH is potentiated by the enhancement of the cholinergic activity but much more by ARG, which is compatible with the presence of a cholinergic impairments.

Spontaneous GH secretion as well as GH response to several stimuli including GHRH have been shown to be reduced in obesity. To clarify the pathogenesis underlying these alterations, in six obese patients (3 males and 3 females, age 20-44 yrs, BMI = 42.1 +/- 2.2) on unrestricted diet we studied the effect of 8 day GHRH pretreatment (1 micrograms/kg iv each day) on the acute somatotropic response to the neurohormone administered both alone and combined with arginine (ARG, 0.5 g/kg iv infused from 0 to 30 min) which likely inhibits the release of hypothalamic somatostatin. Before treatment the GH response to GHRH (AUC: 231.9 +/- 106.4 micrograms/l/h) was potentiated (p < 0.001) by ARG (932.6 +/- 166.2 micrograms/l/h). However, the GH responses to the neurohormone both alone and combined with ARG were lower (p < 0.02 and 0.002, respectively) than in normals (712.4 +/- 111.6 and 2608.3 +/- 453.2 micrograms/l/h, respectively). After repetitive GHRH administration, in obese subjects baseline GH and IGF-I levels were unchanged. Also the GH responses to GHRH both alone (217.3 +/- 68.1 micrograms/l/h) and combined with ARG (756.3 +/- 202.9 micrograms/l/h) were not modified. In conclusion, our data demonstrate the failure of GHRH pretreatment to improve the somatotrope hyporesponsiveness to GHRH both alone and combined with ARG suggesting the existence of a somatotropic defect in obesity.

In insulin dependent diabetes mellitus (IDDM) either elevated growth hormone (GH) levels or abnormal responses to specific as well as unspecific stimuli have been reported. As hyperglycemia is known to blunt GH response to various stimuli, a normal GH response to GHRH in presence of hyperglycemia should also be considered inappropriate. To investigate the mechanism underlying this inappropriate GH response, in 9 patients with IDDM, selected for normal GH response to GHRH, we studied the GH response to two consecutive GHRH boluses (1 microgram/kg), the second of which preceded 30 min before by pyridostigmine (120 mg p.o.). Seven age matched normal volunteers were evaluated as control group. Basal plasma glucose and serum GH levels were significantly higher in patients with IDDM than in normal subjects (184.4 +/- 9.6 vs 86.2 +/- 4.4 mg/dl, p < 0.01 and 2.4 +/- 1.0 vs 1.0 +/- 0.4 microgram/l, p < 0.01 respectively). Both in normal subjects and in patients with IDDM the GH response to the second consecutive GHRH administration was lower than that of the first GHRH bolus (delta AUC: 82.5 +/- 28.3 vs 401.1 +/- 131.2 micrograms/l/h, p < 0.05 and 77.2 +/- 30.4 vs 336.8 +/- 60.0 p < 0.02, respectively). Pyridostigmine was able to restore the blunted GH responsiveness to the second GHRH administration in both groups, but this response was found higher in normal than in diabetic subjects (delta AUC: 1250.8 +/- 136.2 vs 527.5 +/- 147.6, p < 0.01). Since the GH-releasing effect of PD is likely to be mediated by the inhibition of hypothalamic somatostatin release, our results suggest that there is also an impaired somatostatin tone in hyperglycemic type 1 diabetic patients with normal GH response to GHRH.