Ghrelin and growth hormone: story in reverse.

T he current epidemic of obesity and diabetes in the face of a surfeit of calories contrasts the natural selection to survive famine that confronted our ancestors. The report in PNAS by Zhao et al. (1) shows that ghrelin is an important hormone in this process. By studying mice under severe caloric restriction, they show that knockout (KO) of the enzyme ghrelin O-acyltransferase (GOAT), necessary to convert ghrelin to its active form, manifests the first clear phenotype seen in KO of ghrelin activity; ghrelin is necessary for triggering the growthhormone (GH) response to nutritional deprivation that prevents hypoglycemia and death. The road to ghrelin’s discovery started with the observation that morphine stimulates GH secretion, isolation of the enkephalins, and development of enkephalin analogs that selectively stimulate secretion of GH [GH-secretagogues (GHSs)] and orally active GHS. One such GHS, MK-0677, was used to expression clone the GHS receptor (GHS-R) (2). Using cell lines expressing this receptor, Kojima et al. (3) isolated and characterized its endogenous ligand (ghrelin) from rat stomach extracts. They discovered that the third amino acid of ghrelin had an unexpectedly large mass; it was a serine residue modified by the attachment of octanoate, a medium-chain fatty acid. Removing the octanoate blocked ghrelin’s ability to bind to or activate its receptor. This acylation of a peptide with a mediumchain fatty acid is unique to ghrelin. Although ghrelin was named for its GH-releasing activity, its pleiotropic effects, including increased appetite, altered gastrointestinal motility, and regulated lipid and glucose metabolism, cardiac function, blood pressure, immune function, cell proliferation, sleep, anxiety, and even memory, soon became apparent (4). However, KO of ghrelin or GHS-R or the combined KO of both receptor and ligand had no distinctive phenotype (5). The reason that no phenotype was observed with these KO animals is likely because previous studies did not employ the prolonged (10 days) severe nutritional restriction (40% of normal levels) that is used in the study of Zhao et al. (1). Ghrelin’s potent orexigenic properties led a number of groups to suggest that ghrelin may be important in the pathogenesis of obesity, type 2 diabetes, and Prader–Willi syndrome (6, 7). Recently, two groups independently identified GOAT, which had been previously known as an orphan member of a family of membrane-bound Oacyltransferase enzymes (MBOATs) (8, 9). GOAT is expressed in the ghrelinproducing cells of the stomach and small intestine. GOAT is required for the attachment of octanoate to preproghrelin and has no other known activity (8, 9). The report by Zhao et al. (1) combines molecular biological approaches with an integrative biological experimental paradigm. They study a GOAT KO mouse that, therefore, lacks acyl-ghrelin and expose it to prolonged nutritional restriction (40% of normal levels). They show that ghrelin is important for the maintenance of the blood–glucose levels needed for survival during prolonged nutrient restriction. The depletion of fat reserves after more than 3 days of nutritional restriction with feeding one time per 24 h results in debilitating hypoglycemia at the end of the day in the GOAT KO mice but not wild-type mice. Thus a clear metabolic phenotype has been demonstrated in the GOAT KO mouse. Thus, ghrelin is critical in the counterregulatory ensemble of factors that maintains blood glucose (Fig. 1). This repertoire includes catecholamines, glucagon, glucocorticoids, and GH. Earlier studies suggested that ghrelin might increase insulin resistance and impair insulin secretion (10). Zhao et al. (1) define a physiological role for ghrelin; it orchestrates the enhanced GH secretion induced by prolonged nutritional restriction. It must be emphasized that these mice were calorie restricted but not fasted. This is important, because the lipid group that is attached to preproghrelin by GOAT is likely derived from free fatty acids (FFA) in the lumen of the gut rather than circulation (11). Indeed, with prolonged fasting (>37.5 h) in humans, ghrelin levels are suppressed, whereas desacylghrelin is tonically secreted (12). If the FFA required for activation of ghrelin must come from the gut, then ghrelin will be suppressed in wild-type mice during long-term fasting; this contrasts the increased levels described by Zhao et al. (1) during nutritional restriction. The evidence presented suggests that GH is pivotal in preserving normal blood–glucose levels. This is supported by the lower levels of GH observed in the GOAT KO mice and the rescue of blood glucose by infusion of GH or ghrelin. This emphasizes the physiological importance of direct effects of GH in the severely nutritionally restricted state, a time when the levels of insulin-like growth factor I (IGF-I) are low. GH levels are raised in starved infants (at a time when glucose may be low) with profoundly low IGF-I levels, and this reverses with Ghrelin