Ghrelin, a peptide of 28 amino acids, was first identified by Kenji Kangawa and colleagues at the National Cardiovascular Center Research Institute, Osaka, Japan, as the natural ligand for growth hormone secretagogue (GHS) receptors (see Kojima et al., 1999). Back then, the small synthetic GHSs were known to bind to a G protein-coupled receptor that had no known natural ligand. Ghrelin turned out to be able to enter the brain, stimulate eating as well as growth hormone release from the pituitary, and reduce fat utilization. In short, ghrelin revealed a new orexigenic (appetite-increasing) pathway that acted independently of growth hormone-releasing hormone.

But much as ghrelin caused excitement in the field of metabolism research, it right away appeared to be about more than appetite. Both ghrelin and its receptor occur in a variety of tissues, including heart and lungs, suggesting a role in the cardiovascular system. Later, the metabolic action came in for scrutiny when researchers reported that ghrelin-null mice are of normal weight (see Wortley et al., 2004). In parallel, data showing that GHS receptors are abundant in the hippocampus (see Guan et al., 1997), together with data showing that injecting ghrelin into rats improves their memory (see Carlini et al., 2002), began to create a sense that ghrelin may actually be more about learning and memory than yearning and gastronomy (see also review by Ghigo et al., 2005).

Now, Tamas Horvath and colleagues at Yale University, New Haven, Connecticut, and elsewhere, provide convincing data that ghrelin does indeed act on the hippocampus to increase memory performance. In the present paper, lead author Sabrina Diano and colleagues confirmed that those hippocampal receptors actually bind the peptide. The researchers found that in hippocampal slices, neurons in the dentate gyrus, CA1, and CA3 regions readily bind ghrelin, and that the binding colocalizes with the receptor. Using radiolabeled ghrelin, the researchers also confirmed that the peptide crosses the blood-brain barrier from the periphery and enters the hippocampus in vivo.

But what does ghrelin do once it gets there? Horvath and colleagues recently found that the gut hormone can alter synaptic organization in the hypothalamus (see Pinto et al., 2004), so they tested to see if the same might hold true in the hippocampus. The researchers found that after mice received a peripheral injection of the hormone, the number of spine synapses in the CA1 subfield of the hippocampus increased by about 30 percent. Endogenous ghrelin seems to have the same effect because the authors detected significantly fewer spine synapses in the respective regions in ghrelin-null mice.

To determine the significance of the increased synapses, Diano and colleagues first looked at long-term potentiation (LTP), a type of synaptic plasticity that is essential for learning and memory. They found that the peptide promoted LTP when added to hippocampal slices. To check for physiological significance, the researchers gave ghrelin to mice, then evaluated them in various tests of learning and memory. A single injection of the hormone or the GHS receptor agonist LY44711 led to markedly improved performance in the plus-maze test, in which the animals explore, learn, and remember the arms of a maze. Also, in tests where the animals must learn to move to safety to avoid a mild foot shock, ghrelin administered directly to the brain after—rather than before—the animals were trained in the task, led to significantly improved performance one week later. This experiment suggests that the hormone helps the brain consolidate memories.

The effects of endogenous ghrelin also seem to be important for learning and memory because in an object recognition test, where mice are exposed to both familiar and new items, ghrelin-null mice did poorly in comparison to normal animals. This suggests that animals fail to recognize familiar objects in the absence of the hormone.

Perhaps most interestingly for students of Alzheimer disease, the authors found that in the P8 strain of senescence-accelerated mouse (SAMP8), ghrelin improved memory performance in both 12- and 14-month-old animals. SAMP8 mice mimic some of the pathology and symptoms of AD. As they age, they accumulate amyloid-β and show progressive impairments in learning and memory (see Morley et al., 2000).

Though these findings suggest that ghrelin might help tackle some of the symptoms of AD, don’t go on a starvation diet yet. Recent research has shown that weight loss late in life is associated with AD and may start before the disease is clinically obvious (see Stewart et al., 2005 and related comments). Whether this weight loss is a consequence of an underlying disease process or contributes to it is uncertain. Equally uncertain at this point is the question of what the underlying molecular pathways are by which obesity intersects with dementia. Staying slim, trim, and active remains a good recipe for mind and body, but one way to boost ghrelin levels late in life that’s worth exploring may be to take it in pill form. “Orally active GHS analogs may offer a potential treatment for impaired learning and memory processing, which occurs in association with aging and neurological disorders such as Alzheimer disease,” write the authors.—Tom Fagan.

Reference:
Diano S, Farr SA, Benoit SC, Mcnay EC, da Silva I, Horvath B, Gaskin FS, Nonaka N, Jaeger LB, Banks WA, Morley JE, Pinto S, Sherwin RS, Xu L, Yamada KA, Sleeman MW, Tschöp MH, Horvath TL. Ghrelin controls hippocampal spine synapse density and memory performance. Nat Neurosci. 2006 Mar;9(3):381-8. Epub 2006 Feb 19. Abstract