Dark-field image showing ghrelin cells and processes in the hypothalamus. Courtesy of Tamas Horvath, Yale University Medical School, New Haven, Connecticut, USA.

The elusive source of endogenous ghrelin in the hypothalamus is described by Horvath, Cowley and colleagues in a report in Neuron. Ghrelin is a peptide hormone that is produced mainly by the stomach in response to fasting. It activates the growth hormone secretagogue receptor and has powerful actions on feeding and energy regulatory circuits, including a direct influence on hypothalamic neurons. The new results indicate that at least some of these effects might be mediated by brain-derived ghrelin, rather than by ghrelin from the stomach.

Cowley et al. showed that ghrelin is expressed in a previously uncharacterized group of neurons in the hypothalamus. The neurons lie in the space between the lateral hypothalamic, arcuate, ventromedial, dorsomedial and paraventricular hypothalamic nuclei, and they send projections to several of these nuclei as well as outside the hypothalamus. Interestingly, the area that contains the ghrelin neurons overlaps with the projections from the suprachiasmatic nucleus, which might allow the production of ghrelin to be directly modulated by the circadian clock.

Although there have been previous indications that ghrelin is produced in the hypothalamus, the function of this brain-derived ghrelin is unclear. Cowley and colleagues have provided some clues to what it might be doing. Anatomically, they saw that ghrelin-containing boutons are often closely apposed to the axon terminals of other hypothalamic neurons, particularly those containing neuropeptide Y (NPY), raising the possibility that it might act presynaptically to modulate the release of NPY and GABA (γ-aminobutyric acid) by these neurons.

The authors went on to investigate the physiological effects of ghrelin in two hypothalamic nuclei, the arcuate nucleus and the paraventricular nucleus. They found that treatment with ghrelin increases the activity of the NPY-containing neurons to which the ghrelin neurons project, and hyperpolarizes pro-opiomelanocortin (POMC)-containing neurons in the arcuate nucleus. This hyperpolarization is probably mediated by increased release of NPY and GABA.

In the paraventricular nucleus, ghrelin reduced the GABA-mediated inhibition of 73% of neurons, but did not affect the rest. Previous work showed that NPY had the same effect, and the authors propose that the influence of ghrelin here is also mediated by a presynaptic increase in NPY release, which might in turn act presynaptically to reduce GABA secretion.

The existence of an endogenous ghrelin system in the hypothalamus might solve some of the problems that have been associated with the idea that ghrelin from the stomach acts on these nuclei to regulate energy homeostasis. If more work confirms the functions of the ghrelin-containing hypothalamic neurons, it will represent an important step in our understanding of how complex interactions between the brain and the gastrointestinal system regulate our food intake and energy expenditure.