The search for neuroprotective treatments after ischemic brain injury has so far proved remarkably unsuccessful. Indeed, recombinant tissue plasminogen activator (rt-PA), first introduced in the 1980s [1], is currently the only useful treatment for ischemic stroke. Although rt-PA effectively restores blood flow to the brain it cannot directly target the fundamental mechanisms of ischemic injury. Furthermore, owing to the narrow therapeutic window (less than 4.5 h), only a small proportion of stroke patients actually receive rt-PA [2]. With ischemic stroke being the second leading cause of death and disability worldwide [3], there is a clear need for more effective therapies. As will be discussed below, emerging evidence suggests that ghrelin may be one such potential therapy.

Ghrelin is a 28 amino acid peptide that is principally synthesized in the gut, but is also expressed in a variety of other tissues [4,5]. It was first identified in 1999 as a stimulator of growth hormone release [5]. Ghrelin exists in the plasma in a des-acylated and an acylated form, the latter of which is the result of post-translational octanoylation of pro-ghrelin by the enzyme ghrelin-O-acyltransferase (GOAT) [6]. Acylated ghrelin is the natural ligand of the growth hormone secretagogue receptor type 1a (ghrelin receptor, GHS-R1a) through which growth hormone release is stimulated [5,6,7].

Since this initial description, research has attributed many other functions to the peptide, including its well-known role in feeding and metabolism [8,9,10,11]. Under situations of food deprivation, acylated ghrelin signals through the hypothalamus to promote feeding and energy conservation [10,12]. Ghrelin also has a number of significant extra-hypothalamic neuronal functions, including in reward and motivation [13,14], learning and memory [15], and stress [16,17]. For instance, we have recently shown ghrelin attenuates the hypothalamic-pituitary-adrenal axis response to acute stress, probably by acting on the anterior pituitary to stimulate adrenocorticotropic hormone release and thus enhance glucocorticoid negative feedback [16].

The GHS-R1a is expressed throughout the brain including in feeding and metabolism-associated areas like the arcuate nucleus (ARC) [18,19,20], as well as regions important in memory (Cornu Ammonis (CA)2, CA3, dentate gyrus of the hippocampus) and reward (ventral tegmental area, substantia nigra pars compacta (SNpc), dorsal raphe) [19,21,22]. The GHS-R1a is a g-protein coupled seven transmembrane receptor. Ghrelin activates the GHS-R1a leading to stimulation of the phospholipase C (PLC)/protein kinase C (PKC)/inositol trisphosphate (IP3) pathway that triggers IP3-dependent calcium release from intracellular stores. This intracellular calcium coupled with calcium entering the cell via voltage-gated L-type calcium channels stimulates the GHS-R1a’s downstream responses [23,24,25]. In addition to its ghrelin-dependent effects, the GHSR has high constitutive activity in the absence of the ligand [26]. It is also able to dimerize with other receptors such as the dopamine receptor subtype 2 to modulate dopamine signaling [27], and the melanocortin-3 receptor to modulate melanocortin signaling [28]. Ghrelin itself is also able to cross the blood brain barrier [5,18,29,30], making the ligand and receptor ideally placed to play an integral role in these diverse functions. As discussed below, evidence now suggests ghrelin may also have a role in neuroprotection in the setting of neurodegenerative disease and ischemic injury.

Experimental evidence in animal models clearly suggests ghrelin may be a useful therapeutic after cerebral ischemia. This role may prove to be all the more crucial when we consider the ghrelin system appears to be altered in the setting of known stroke risk factors such as aging, obesity, and hypertension. For example, during the normal aging process there is a reduction in the sensitivity to the metabolic effects of ghrelin [76,77,78]. This insensitivity may be compounded by an aging-related decline in circulating ghrelin levels [39]. Obesity is associated with a central resistance to the metabolic effects of ghrelin [79], and ghrelin levels are reduced in obese and overweight humans [80]. Similarly, low ghrelin levels have also been reported to be independently associated with the development of type two diabetes and elevated blood pressure [81], and the Arg51Gln (rs34911341) single-nucleotide polymorphism of ghrelin is linked to an increased risk of hypertension, and is associated with low ghrelin levels [82]. Circulating ghrelin is also lower in patients who have experienced stroke than in the general population when controlling for factors such as age and obesity [47]. Collectively, these findings raise the intriguing possibility that low ghrelin levels per se may be a risk factor for stroke, and imply that ghrelin might not only be an effective intervention therapy after a stroke has occurred, but also an approach to prevent a stroke from occurring in the first place. There is therefore a strong rationale to evaluate whether increasing ghrelin levels and/or improving sensitivity to circulating ghrelin during aging is an effective strategy to prevent ischemic stroke damage. Of importance, experimental evidence indicates ghrelin can influence the sensitivity of the brain to its actions [62,83]. Specifically, studies have shown that GHS-R1a expression in the brain is increased following either central or peripheral administration of ghrelin to rodents [62,83]. Thus, sensitivity to circulating ghrelin could be improved during stroke risk factors (and after stroke) by “simply” increasing circulating ghrelin levels through ghrelin supplementation.

An important consideration with using ghrelin as a therapeutic is the timing of the dose. As suggested, restoring normal ghrelin levels in aging or other populations where ghrelin is low may be a viable therapy, and in this case chronic ghrelin treatment would be necessary. An alternative strategy is to employ a single or repeated supra-physiological dose as a therapeutic agent and the vast majority of preclinical studies have tested single doses of exogenous ghrelin at the time of injury (Table 1). Unfortunately, this is likely to be effective only in the 30 min or so after administration, as ghrelin’s half-life is very short (approximately 25–30 min in humans [84,85]). As such, a single dose at this time would likely influence early apoptosis and inflammation, but not affect the ongoing ischemic injury. Since brain damage after ischemic injury can continue to worsen for hours to days and weeks after the event depending upon the type of injury [86], chronic ghrelin treatment may be more appropriate after injury as well as a prophylaxis. In either case, such treatments need to be considered carefully and further research is needed. For instance, studies have shown glutamatergic neurotransmission is necessary for ghrelin’s effects on mesolimbic dopaminergic reward pathways [87] and that ghrelin stimulates memory formation in a glutamate-dependent manner [15]. However, excessive glutamate release can lead to excitotoxicity [88], which would obviously be detrimental after ischemic injury. The circadian timing of the ghrelin administration must also be considered, as evidence from the PD model suggests exogenous ghrelin may only be effective if the subject is fasted [36].

When evaluating the overall usefulness of ghrelin as a therapeutic for cerebral injury after ischemia, it is also important to consider the potential long-term consequences on body homeostasis. For example, exogenous acylated ghrelin would be expected to act at the GHS-R1a to stimulate appetite, food intake, and weight gain. While this may be a problem in the general population, it may actually be of benefit in those recovering from stroke as weight loss after stroke has been linked with a worse long-term prognosis [89]. Thus, ghrelin as a stroke therapy could have an additional beneficial effect by preventing age-related malnutrition and frailty, as well as weight loss after stroke. An alternative option in those where weight gain is a concern is also to employ the des-acylated form of ghrelin. Des-acylated ghrelin does not act at GHS-R1a and does not have the same metabolic effects as the acylated form. It is, however, effective at preventing neurodegeneration and improving outcomes after ischemic damage [56]. Further research will be necessary to determine its mechanism of action and effects on other neuronal functions such as reward and stress, but treatment with des-acylated ghrelin is nonetheless a useful potential option.

The research reviewed here suggests the peptide hormone, ghrelin, may be an exciting novel candidate in our search for treatments for ischemic brain injury. It is clear ghrelin treatment can provide significant neuroprotection in a number of models of neurodegenerative disease and following ischemic brain injury. Indeed, ghrelin is effective in improving cell survival, reducing infarct size and rescuing memory in these models. It does this primarily by suppressing the apoptosis and inflammation associated with the ischemic injury. However, it remains to be determined whether ghrelin also modulates necrotic or autophagic mechanisms of cell death. Nonetheless, these exciting data collectively point to ghrelin potentially being very useful in the clinic as a neuroprotective treatment.