Ghrelin Gene Products in Acute and Chronic Inflammation
- First Online:
- Cite this article as:
- Prodam, F. & Filigheddu, N. Arch. Immunol. Ther. Exp. (2014) 62: 369. doi:10.1007/s00005-014-0287-9
- 1.2k Downloads
Ghrelin gene products—the peptides ghrelin, unacylated ghrelin, and obestatin—have several actions on the immune system, opening new perspectives within neuroendocrinology, metabolism and inflammation. The aim of this review is to summarize the available evidence regarding the less known role of these peptides in the machinery of inflammation and autoimmunity, outlining some of their most promising therapeutic applications.
KeywordsAcylated ghrelin Unacylated ghrelin Obestatin Autoimmune disease Inflammatory disease
Anti-neutrophil cytoplasmic antibody
Body mass index
Chronic obstructive pulmonary disease
[d-Lys3] Growth hormone releasing peptide-6
Growth hormone secretagogue receptor type 1a
Glycogen synthase kinase
Tumor necrosis factor-alpha
Vascular cell adhesion molecule-1
Ghrelin is a circulating acylated peptide that promotes a strong release of growth hormone (GH) through binding to and activation of its receptor GH secretagogue receptor type 1a (GHS-R1a) in hypothalamus and pituitary (Kojima et al. 1999; van der Lely et al. 2004). In recent times, ghrelin gained popularity as the “hunger hormone” because its secretion, mainly from the stomach, is regulated by fasting and, in turn, it induces appetite, promotes adiposity, and controls energy homeostasis (Ariyasu et al. 2001; Tschöp et al. 2000; Wren et al. 2001). In humans, ghrelin secretion is pulsatile, with a higher secretion during the night. Plasma concentration of ghrelin rises during fasting, drops to basal values after the assumption of food, and it could contribute to meal initiation or choose of nutrients (Cummings 2006). Ghrelin is commonly altered in pathological conditions affecting body mass and/or body energy metabolism, negatively correlating with body mass index (BMI) (van der Lely et al. 2004). Ghrelin circulating levels are indeed lower in overweight or obese subjects than in normal individuals (Tschöp et al. 2000), with the exception of Prader Willi syndrome (Cummings et al. 2002), and higher in conditions characterized by energy inadequacy such as anorexia/cachexia associated with cancer (Garcia et al. 2005; Shimizu et al. 2003) and several other pathologies including chronic heart and renal failure, chronic obstructive pulmonary disease (COPD), and cystic fibrosis (Cohen et al. 2008; Itoh et al. 2004; Nagaya et al. 2001; Yoshimoto et al. 2002). Insulin and glucose are among the major determinants of ghrelin secretion that, in turn, modulates insulin secretion and glucose metabolism (van der Lely et al. 2004; Wiedmer et al. 2007). Recently, the specific acyltransferase that octanoylates ghrelin (GOAT: ghrelin O-acyltransferase) was detected in human circulation in healthy, obese, and anorexic adults with a positive correlation with BMI and a negative correlation with ghrelin levels (Goebel-Stengel et al. 2013).
Beside its role in feeding and energy homeostasis, ghrelin exerts also an extensive array of other biological activities such as cardio-protection and the enhancement of cardiac function (Baldanzi et al. 2002; Nagaya et al. 2004), pro-kinetic actions on gastric motility (Chen and Tsai 2012), and neuronal activities (Raimondo et al. 2013), including those on sleep (García-García et al. 2014), epilepsy (Portelli et al. 2012), and learning and memory (Beck and Pourié 2013).
The unique acylation of ghrelin on serine 3, mainly with octanoic acid, is essential for ghrelin activity through GHS-R1a. Indeed, the unacylated form of ghrelin, which accounts for most of the circulating peptide, does not bind to and activate GHS-R1a (Kojima et al. 1999), unless used at micromolar concentrations, because at such high concentrations the unacylated form can act as a functional agonist (Bednarek et al. 2000; Gauna et al. 2007; Heppner et al. 2014; Matsumoto et al. 2001). However, starting with the first demonstration in vitro of the anti-apoptotic activity of unacylated ghrelin on cardiomyocytes and endothelial cells (Baldanzi et al. 2002), and, then, in vivo on insulin secretion in humans (Broglio et al. 2004), a continuously increasing number of works demonstrated that unacylated ghrelin is a biologically active peptide, sharing, often in cells lacking GHS-R1a, most of acylated ghrelin biological activities.
Among the wide range of common biological functions, acylated and unacylated ghrelin exert a protective activity on several cell lines including cardiac and endothelial cells, pancreatic β cells, islets, and islet microendothelial cells, cortical neurons, vascular smooth muscle cells, and visceral adipocytes (Baldanzi et al. 2002; Chung et al. 2008; Favaro et al. 2012; Granata et al. 2007; Hwang et al. 2009; Rodríguez et al. 2012; Zhan et al. 2008). Both peptides promote the differentiation of skeletal myoblasts, preadipocytes, and embryonic stem cells toward cardiomyocytes (Filigheddu et al. 2007; Gao et al. 2012, 2013; Giovambattista et al. 2008; Miegueu et al. 2011). Moreover, they affect, either positively or negatively, proliferation of numerous cell types including osteoblasts, pancreatic β cells, islets and islet microendothelial cells, myoblasts, and adrenocortical tumor cells (Delhanty et al. 2006, 2007; Favaro et al. 2012; Filigheddu et al. 2007; Granata et al. 2007).
Since unacylated ghrelin shares with ghrelin common high affinity binding sites in various cell lines, including cells lacking GHS-R1a (Baldanzi et al. 2002; Cassoni et al. 2001, 2004; Filigheddu et al. 2007; Granata et al. 2007; Jeffery et al. 2002; Muccioli et al. 2004), and they share biological activities in vivo, also in Ghsr null mice (Porporato et al. 2013), both peptides act in all probability through a common, although yet unidentified, receptor.
In some conditions, however, unacylated ghrelin counteracts ghrelin activities both in vitro and in vivo (Benso et al. 2012; Broglio et al. 2004, 2008; Gauna et al. 2005), suggesting a complex interplay between the two forms of the peptide. Unacylated ghrelin participates also in the regulation of food intake and adipogenesis through mechanisms not fully elucidated, but very likely independent of GHS-R1a (Asakawa et al. 2005; Inhoff et al. 2008; Toshinai et al. 2006).
The processing of the 117 aa peptide encoded by the ghrelin gene results, in addition to acylated and unacylated ghrelin, in a 23 aa peptide named “obestatin” for its supposed anti-obesity activity achieved by contrasting ghrelin effects through the binding to GPR39 receptor (Zhang et al. 2005). Successive works, however, did not support these assertions (Bassil et al. 2007; Chartrel et al. 2007; Lauwers et al. 2006), and the biological role of obestatin is still debated (Lacquaniti et al. 2011). Nevertheless, obestatin has several biological activities, including protective effects on the cardiovascular system (Agnew et al. 2012; Alloatti et al. 2010; Aragno et al. 2012), promotion of adipocytes survival, differentiation, and metabolism (Granata et al. 2012; Gurriarán-Rodríguez et al. 2011; Miegueu et al. 2011; Pruszynska-Oszmalek et al. 2013), induction of myoblasts differentiation and promotion of skeletal muscle regeneration (Gurriarán-Rodríguez et al. 2012), inhibition of pancreatic β-cell and islets apoptosis, stimulation of glucose-stimulated insulin secretion in both normal and diabetic animal models, and enhancement of functional β-cell generation from pancreatic precursor cells (Baragli et al. 2013; Granata et al. 2008, 2010). Notably by an evolutionary point of view, while ghrelin is mainly involved in food intake, obestatin has an inhibitory effect on water drinking by acting on mechanisms involved in the thirst (Samson et al. 2007) and decreases vasopressin concentration in plasma (Samson et al. 2008).
Circulating obestatin concentrations have been reported to be either reduced (Anderwald-Stadler et al. 2007; Nakahara et al. 2008) or, more frequently, increased (Guo et al. 2007; Prodam et al. 2011; Reinehr et al. 2008; Vicennati et al. 2007) in obesity. Based on available data, it is difficult to explain these contrasting findings that seem to depend mainly on sample size, age, and ethnicities. Obestatin is also regulated by fasting and food intake, although with different kinetics with respect to acylated and unacylated ghrelin (Guo et al. 2008; Prodam et al. 2011). It is partially refractory to the inhibition by the exogenous administration of insulin (Anderwald-Stadler et al. 2007), as also to the physiological increase in insulin at the glucose challenge (Prodam et al. 2011).
Ghrelin and GHS-R1a mRNA Expression and Distribution in the Lymphoid System
Real-time PCR analysis revealed the expression of ghrelin mRNA in human T cells (Dixit et al. 2004; Hattori et al. 2001), B cells, neutrophils, macrophages, monocytes, dendritic cells, natural killer cells, and several leukemic cell lines (Baatar et al. 2011; Delgado and Ganea 2008; Dixit et al. 2004; Hattori et al. 2001). In addition, when activated, T cells also express the preproghrelin protein and produce and secrete both acylated and unacylated ghrelin (Baatar et al. 2011; Dixit et al. 2004). Besides lymphocytes, ghrelin mRNA was also found in human spleen, lymph nodes, and thymus (Dixit et al. 2004; Gnanapavan et al. 2002). In thymus, ghrelin and the GHS-R1a decline with age. Ghrelin seems to be a novel regulator of epithelial-to-mesenchymal transition and preserves thymic stromal cell microenvironment by controlling age-related adipocytes development within the thymus (Youm et al. 2009).
Also GHS-R1a is widely expressed in the lymphoid system, and its expression was reported in the same cells expressing ghrelin (Dixit et al. 2004; Hattori et al. 2001), as well as in the spleen (Gnanapavan et al. 2002). A truncated inactive GHS-R1b has been shown in various immune tissues, although ghrelin does not bind to and activate it and the role of this receptor is still unknown (Taub 2008). Expression of ghrelin and GHS-R1a was detectable in all immune cells regardless of the maturity and cell types with huge variations due to inter-individual changes and to the activation state of the cells being examined (Baatar et al. 2011; Hattori et al. 2001).
Activation of T cells results in the increase in the GHS-R1a expression and its localization at the plasma membrane in lipid rafts. The same localization of endogenous ghrelin upon T cell activation suggests its possible autocrine/paracrine function in immune cells (Dixit et al. 2004).
The wide expression of ghrelin and GHS-R1a in the lymphoid system strongly suggests a role for this peptide and its receptor in the regulation of the immune system activity, but ghrelin activity could be even wider than suspected. [d-Lys3]-GHRP-6 is a synthetic peptide used in vitro and in vivo as a selective GHS-R1a antagonist, despite the fact that in some conditions it acts as a functional agonist of the GHS-R1a (Depoortere et al. 2006; Erriquez et al. 2009), and in others it is not able to counteract the endocrine functions of ghrelin (Benso et al. 2007). Recently, [d-Lys3]-GHRP-6 has been shown to act as antagonist on the CXCR4 and CCR5 chemokine receptors in T cells and peripheral blood mononuclear cells (Patel et al. 2012a, b). However, no data are still available on ghrelin activity on these receptors. Notably, GHS-R1a is activated also by another neuropeptide, cortistatin, which, although encoded by a different gene, shares high homology with somatostatin and binds to all the known somatostatin receptor subtypes with similar affinity and to MrgX2, a previously orphan G-protein-coupled receptor expressed in the dorsal root ganglions. The binding to GHS-R1a and to MrgX2 is highly specific for cortistatin, being somatostatin devoid of it (de Lecea and Castaño 2006). The ability of cortistatin to bind to GHS-R1a is particularly intriguing, since somatostatin and its fragments do not bind all the same receptor (de Lecea and Castaño 2006). Human immune cells produce cortistatin but not somatostatin. Macrophage lineage, activated endothelium, human lymphocytes and thymus express several somatostatin receptor subtypes and the MrgX2 receptor (van Hagen et al. 2008), suggesting that cortistatin might represent the link between ghrelin and the somatostatin/cortistatin system in the immune system.
Ghrelin, Unacylated Ghrelin, and Obestatin Secretion and Activity in Acute and Chronic Inflammatory States
Several studies on the anti-inflammatory activity of ghrelin showed that only the acylated peptide was able to decrease the expression and production of proinflammatory cytokines: interleukin (IL)-1α, IL-1β, IL-6, and tumor necrosis factor (TNF)-α by activated T cells and by lipopolysaccharide (LPS)-activated monocytes and microglia (Dixit et al. 2004; Li et al. 2004; Theil et al. 2009). Ghrelin has been shown to prevent also microglial activation in a mouse model of Parkinson’s disease by inhibiting the expression of TNF-α and IL-1β induced by administration of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) and to prevent the release of TNF-α and IL-1β from microglia treated with threohydroxyaspartate (Lee et al. 2012; Moon et al. 2009). Since GHS-R1a is not expressed in microglial cells, this effect of ghrelin on microglia activation might be either indirect or could be mediated by a yet unknown receptor. Experiments carried out in vitro on N9 microglia cells, which do not express GHS-R1a, produced quite different results, since only unacylated ghrelin, but not acylated ghrelin, was able to impair the expression of IL-1β, IL-6 induced by exposure to fibrillar beta-amyloid protein (Bulgarelli et al. 2009).
The effect of ghrelin on pro-inflammatory cytokines induction and release has been related, in several studies, to a suppression of the expression and/or activity of the transcription factor NF-κB (Barazzoni et al. 2014; Hou et al. 2009; Konturek et al. 2006; Li et al. 2004; Liu et al. 2010; Peng et al. 2012; Slomiany and Slomiany 2013; Wang et al. 2012; Waseem et al. 2008; Wu et al. 2005; Yuan et al. 2009; Zhou and Xue 2009). However, it has been proposed that in some circumstances and with some cell types, ghrelin may act as a pro-inflammatory peptide inducing, rather than reducing, NF-κB expression and pro-inflammatory IL-8 secretion (Kwan et al. 2010; Rezaeian et al. 2012; Sung et al. 2011; Zhao et al. 2006). Also the vagus nerve seems to play a central role in ghrelin-mediated inhibition of proinflammatory cytokine release (Cheyuo et al. 2011; Wu et al. 2007a, 2008, 2009) and that anti-inflammatory activity of vagal stimulation is ghrelin-mediated (Bansal et al. 2012).
Although unacylated ghrelin is not able to induce the same effects of the acylated peptide in activated T cells and endothelial cells, (Dixit et al. 2004; Li et al. 2004), nevertheless it impairs TNF-α induction in muscles and livers of rats undergoing burn injury and reduces, in myotubes in vitro, the activation of NF-κB induced by treatment with pro-inflammatory cytokines (Sheriff et al. 2012).
Also obestatin has been shown to reduce circulating levels of TNF-α and to impair NF-κB translocation in left ventricular tissue of diabetic rats (Aragno et al. 2012).
Ghrelin, unacylated ghrelin, and obestatin concentrations in acute and chronic inflammatory states and in autoimmune diseases in humans
Acute and chronic inflammatory states
↑ after LPS administration
↑ (in CSF)
Type 1 diabetes
↓ at the onset
↓ or -- after insulin
↓ at the onset
↓ or -- after insulin
↑ or -- in hypothyroidism
↓ in hyperthyroidism
↓ in hyperthyroidism
↑ or -- in hypothyroidism
↓ in hyperthyroidism
↑ at the diagnosis
-- during GFD
Ghrelin physiologically binds to plasma immunoglobulins (Ig) and conditions of altered nutrition affect the affinity of ghrelin for IgG. In particular, it has been suggested that plasma ghrelin-reactive IgG auto-antibodies serve as carrier proteins that might enhance the bioactivity of ghrelin, thus explaining the phenomenon of enhanced orexigenic activity of ghrelin in obesity (Takagi et al. 2013). On the other hand, in anorexia nervosa, these IgG auto-antibodies exist mainly as immune complexes with unacylated ghrelin, leading to a decrease in a free fraction of these auto-antibodies binding acylated and unacylated ghrelin (Terashi et al. 2011) and this may explain the decreased orexigenic effects of ghrelin in this pathology.
Because of the well-known role of ghrelin in feeding and energy homeostasis, many reviews discuss its role with respect to obesity (Prodam et al. 2008; Wiedmer et al. 2007). Therefore, we will discuss other aspects of the regulation and action of ghrelin in inflammatory conditions other than obesity. Moreover, more recently some evidences suggest that unacylated ghrelin may have a therapeutic role in obesity and type 2 diabetes comorbidities. It has been shown that a treatment with both unacylated ghrelin and an analog of it prevented in a murine pre-diabetes model the proinflammatory effects elicited by a high-fat diet. This effect was coupled with an expression of mitochondrial function markers in brown adipose tissue, and with prevention of development of the pre-diabetic metabolic state (Delhanty et al. 2013). Furthermore, systemic administration of unacylated ghrelin prevented diabetes-induced damage of endothelial progenitor cells and facilitated their mobilization from bone marrow (Togliatto et al. 2010).
Cachexia is a multifactorial syndrome, associated with various pathological conditions, and characterized by severe weight loss, muscle and adipose tissue wasting, and inflammation. Ghrelin circulating concentrations are increased in pathological conditions characterized by cachexia, including chronic heart failure, COPD, and different types of cancer (Garcia et al. 2005; Itoh et al. 2004; Kerem et al. 2008; Nagaya et al. 2001; Shimizu et al. 2003). Regardless of the underlying disease, this cachexia-associated hyperghrelinemia may represent either a compensatory mechanism to the negative energy balance and the inflammatory state or the development of a resistance to the ghrelin signaling. This latter hypothesis is supported by the finding that in tumor-bearing cachectic mice the orexigenic effect of ghrelin is reduced compared to non-cachectic littermates (Wang et al. 2006).
Sepsis, the systemic inflammatory response to an infection, may be associated with dysfunctions of several organs, including heart, kidneys, liver, and lungs. Several studies, both in animal models and humans, have demonstrated that sepsis modulates plasmatic ghrelin concentrations and that the administration of ghrelin has beneficial effects on sepsis. In surgical patients with postoperative intra-abdominal sepsis plasma levels of ghrelin increased and positively correlated with TNF-α and IL-6 levels (Maruna et al. 2005). In healthy human subjects, LPS administration induced a biphasic change in circulating ghrelin levels, with a rapid increase reaching the maximum after the pick of TNF-α and a subsequent decrease with the nadir 5 h after LPS administration (Vila et al. 2007). In rodent models of sepsis, induced either by LPS injection or caecal ligation and perforation, ghrelin treatment reduced the level of pro-inflammatory cytokines both in circulation and in tissues impacted by sepsis, such as lung and kidney (Peng et al. 2012; Wang et al. 2009; Wu et al. 2007b). It has been shown that ghrelin protective effects are mediated by the activation of the vagal nerve (Wu et al. 2007a), inhibition of sympathetic nervous system, seen as reduced gut-derived norepinephrine release (Jacob et al. 2010; Wu et al. 2007c), and inhibition of NF-κB pathway (Peng et al. 2012; Wu et al. 2007b). In addition, ghrelin exerts its protective action through inhibition of high-mobility group box release (Chorny et al. 2008; Wu et al. 2009) and through an antimicrobial activity (Chorny et al. 2008).
Cystic fibrosis is a genetic disease that arises from mutations in the cystic fibrosis transmembrane conductance regulator gene, which cause a dysfunction of chloride channels. Chronic cachexia and refractory chronic respiratory infections are typical features of the disease. Three week intravenous ghrelin administration suppressed airway inflammation through a decrease in neutrophil titer in lungs of cachectic patients with chronic respiratory infections (Kodama et al. 2008). Regarding the specific regulation of the ghrelin system in cystic fibrosis, total and acylated ghrelin levels were higher in patients in agreement with their cachectic condition (Cohen et al. 2008; Monajemzadeh et al. 2013). GHS-R1a expression decreases in lymphocytes from cachectic cystic fibrosis patients with acute exacerbation. This may only partially be explained as a down-regulation in response to the observed increase in ghrelin circulating levels, since, after antibiotic treatment, GHS-R1a expression in lymphocytes returned to baseline, while ghrelin levels remained increased (Cohen et al. 2010).
Atherosclerosis is a chronic inflammatory disease characterized by thickening and hardening of vessels due to accumulation of lipids. Macrophages play a major role in the development of atherosclerosis, accumulating cholesterol through an increased uptake of oxidized low-density lipoproteins that leads to foam cell formation.
The alteration of circulating levels of acylated and unacylated ghrelin in some typology of patients (Kotani et al. 2006; Pöykkö et al. 2006; Yano et al. 2009) and the increase in radiolabeled ghrelin binding in the atherosclerotic cardiovascular system (Katugampola et al. 2001) strongly suggest a role of ghrelin gene products in this pathology.
Although the anti-inflammatory activity of ghrelin may at once suggest an anti-atherosclerotic activity, the role of this peptide in the development and progression of atherosclerosis appears actually quite controversial. Ghrelin increases the efflux of cholesterol from macrophages through regulation of peroxisome proliferator-activated receptor γ, thus inhibiting the formation of foam cells and reducing the development of atherosclerotic lesions (Cheng et al. 2010; Demers et al. 2009). In TNF-α-treated human endothelial cells, ghrelin inhibits proinflammatory cytokine production, NF-κB activation, expression of adhesion molecules, and adhesion of monocytes (Kellokoski et al. 2009; Li et al. 2004). Diversely, without TNF-α stimulation, ghrelin shows a pro-atherogenic activity (Kellokoski et al. 2009; Skilton et al. 2005). Finally, mice lacking both ghrelin receptor and low-density lipoprotein receptor fed with a high-fat, high-cholesterol diet show no differences in atherosclerotic plaque progression compared with their controls, suggesting that Ghsr is not involved in the plaque development and/or progression (Habegger et al. 2011). However, this does not exclude that at least some of the observed effects of ghrelin on atherosclerosis may be mediated by the alternative, yet unknown ghrelin receptor.
Obestatin reduces adhesion molecule VCAM-1 expression and oxidized low-density lipoprotein binding to macrophages, an early step in foam cells formation (Kellokoski et al. 2009). Moreover, decreased levels of obestatin in subjects with metabolic syndrome associate with an increased carotid atherosclerosis (Cui et al. 2012).
Although it has been suggested that unacylated ghrelin may represent a cardiometabolic marker for predicting atherosclerosis in elderly hypertensive subjects, where its circulating levels are altered and inversely correlated with atherosclerosis markers as carotid intima media thickness (Yano et al. 2009), no in vitro studies have been performed to investigate a possible role for this peptide in the development and/or progression of the pathology.
Alzheimer’s disease is a neurodegenerative disorder characterized by an early hippocampal damage. The pathogenesis is still unknown, although it is well known that the accumulation of amyloid-β peptide and the following chronic microglia inflammation exert a crucial role. Also insulin resistance and the GH/IGF-I system have a functional role in the development of the disease. Because many reports show that ghrelin is involved in neuroprotection, learning, and memory, the assumption of a role in Alzheimer’s disease is fairly justified. First of all, ghrelin, GOAT, and GHS-R1a are all down-regulated in the temporal gyrus, one of the most compromised cortical regions in Alzheimer’s disease (Gahete et al. 2010). This pattern was associated with a shift in the GHS-R1a/GHS-R1b ratio, suggesting an impairment of the autocrine/paracrine mechanisms (Gahete et al. 2010). Conversely, ghrelin increases insulin-stimulated neuronal glucose uptake, decreases insulin resistance, and altered tau phosphorylation via the PI3-K/Akt-GSK pathway in hippocampal neuron cultures (Chen et al. 2010), suggesting that it may represent a potential treatment. Indeed, in murine models of the disease, treatments with ghrelin or GHS-R1a agonists reduced the levels of amyloid-β peptide and inflammation of the microglia (Dhurandhar et al. 2013; Moon et al. 2011). In particular, ghrelin decreases the production of superoxides, mitochondrial membrane depolarization induced by amyloid-β oligomers and prevents GSK-3β activation (Martins et al. 2013). Amyloid-β peptides are able to bind to CD36, a multifunctional protein that is a type B scavenger receptor and that possesses binding sites for the synthetic GHS, like hexarelin (Demers et al. 2004). Hexarelin and unacylated ghrelin, but not acylated ghrelin, seem to interfere with amyloid-β peptide activation of CD36 in microglia cells (Bulgarelli et al. 2009), suggesting that the protective role of the ghrelin system in Alzheimer’s disease may occur, at least in part, in a GHS-R1a independent manner. Interestingly, it has been recently shown that semagacestat, a γ-secretase inhibitor, which belongs to the therapeutic agents for Alzheimer’s disease, is able to activate the GHS-R1a alone or in synergy with acylated ghrelin itself (Schellekens et al. 2013). Because only a few data have been published in patients with Alzheimer’s disease and reported unchanged ghrelin levels in both plasma and cerebrospinal fluid with respect to age-matched controls (Proto et al. 2006; Theodoropoulou et al. 2012), more efforts are needed to better characterize the physiology of the ghrelin system in this condition.
Uveitis is an inflammatory disease of the uveal tract including the iris, ciliary body, and choroidea. It can cause permanent damages in ocular tissues, and sometimes visual loss. Its pathogenesis is still unknown, but several cytokines are involved, among the others IL-1, IL-6, and TNF-α that can be detected in ocular fluids of patients. Because of the anti-inflammatory activity of ghrelin and the presence of ghrelin and GHS-R mRNA in rat and human eyes (Azevedo-Pinto et al. 2013), some authors investigated if ghrelin administration could be beneficial for the treatment of uveitis. However, the intraperitoneal administration of ghrelin at the dose of 10 ng/kg/day for one week, diversely from tacrolimus or infliximab, failed to reduce IL-1, IL-6, and TNF-α in the vitreous of a rat model of uveitis (Gül et al. 2013; Turgut et al. 2013). However, this could be due to an improper route of administration or to a too low dose.
Ghrelin, Unacylated Ghrelin, and Obestatin Secretion and Activity in Autoimmune Diseases
Multiple sclerosis is an autoimmune inflammatory disease characterized by the progressive loss of axons myelination, leading to an altered transmission of nerve signals. Patients, both relapsing-remitting and progressive, have significantly higher circulating levels of total ghrelin than healthy subjects (Berilgen et al. 2005), and a more recent study showed a significant increase in unacylated ghrelin in cerebrospinal fluid that correlated with serum concentrations in patients but not in healthy controls (Unger et al. 2013). These findings suggest that ghrelin may affect the central inflammatory process in multiple sclerosis. Moreover, an association of a haplotype in the GHS-R gene with multiple sclerosis has been reported in some but not all cohorts of patients (Rey et al. 2011), suggesting that the system should be more studied in this disease in the future. Furthermore, in mouse experimental autoimmune encephalomyelitis, a model of multiple sclerosis, sc ghrelin administration was followed by a reduced clinical severity of the disease. This effect was coupled with reduced mRNA levels of proinflammatory cytokines (TNF-α, IL-1β, IL-6) in the spinal cord cellular infiltrates and microglia (Theil et al. 2009). In addition, the ability of ghrelin to decrease the presence and the activation of encephalitogenic T helper 1 (Th1) and Th17 cells in periphery and nervous system and to induce regulatory T cells reinforces the idea of ghrelin as a promising agent to arrest or slow the progression of this disease (Souza-Moreira et al. 2013).
Inflammatory Rheumatic Diseases
Rheumatoid arthritis is a chronic inflammation of the musculoskeletal system, in which TNF-α is the main cytokine involved. Total and acylated ghrelin are similar to controls in adult patients with the disease, whereas interestingly, obestatin correlated with some inflammatory markers, in particular TNF-α and IL-6 (Koca et al. 2008). In agreement with these findings, increased circulating ghrelin levels and higher mRNA expression in the arcuate nucleus were recorded in rats with adjuvant arthritis, which is similar to human rheumatoid arthritis (Skurlova et al. 2010; Stofkova et al. 2009). Higher ghrelin levels were also reported in ANCA-associated vasculitis with a direct association with the severity of the disease (Kümpers et al. 2008).
Long-term treatments with TNF-α inhibitors increased (Gonzalez-Gay et al. 2008) or did not change ghrelin secretion (Toussirot et al. 2013) irrespectively of a significant gain in fat mass or of the presence of metabolic syndrome in patients with rheumatoid arthritis or ankylosing spondylitis. A similar increase after treatment with TNF-α inhibitors was demonstrated also in juvenile idiopathic arthritis (Karagiozoglou-Lampoudi et al. 2011). On the other hand, a transient decrease in total and acylated ghrelin levels in the short-term treatment has been recorded (Chen et al. 2013; Magiera et al. 2013). It can be speculated that a functional hyperghrelinemia, insensitive to the regulation of the fat mass accrual, is a specific characteristic of these inflammatory diseases.
Autoimmune Endocrine Diseases
Type 1 Diabetes
Autoimmune-mediated destruction of β cells causes type 1 diabetes and a decrease in β-cell mass. Most studies reported lower total and acylated ghrelin levels at the onset of the disease (Ashraf et al. 2007; Soriano-Guillén et al. 2004; Holdstock et al. 2004; Martos-Moreno et al. 2006), with a partial recovery of acylated ghrelin levels (Martos-Moreno et al. 2006) and total ghrelin levels in some (Bideci et al. 2005; Holdstock et al. 2004) but not all studies (Ashraf et al. 2007; Celi et al. 2005; Huml et al. 2011; Soriano-Guillén et al. 2004) after a short-term insulin treatment. It has been demonstrated that insulin is essential for meal-induced suppression of ghrelin in adults with type 1 diabetes (Murdolo et al. 2003). Other studies observed a negative correlation between total or acylated ghrelin levels and mean daily insulin dosage in children and adolescents with type 1 diabetes, suggesting that insulin is one of the main regulators of ghrelin secretion in type 1 diabetes (Bideci et al. 2005; Celi et al. 2005; Huml et al. 2011). Because acylated ghrelin is similar to controls after the starting of insulin regimen and higher ghrelin levels are reported in streptozotocin-induced diabetic rats (Dong et al. 2006; Gelling et al. 2004), this suggests that in type 1 diabetes ghrelin is partially refractory to the inhibition by insulin. This finding is in agreement with the hypothesis of the involvement of acylated ghrelin in the hyperphagia and weight gain of type 1 diabetes after the onset of the disease. It has been reported that exogenous administration of acylated ghrelin ameliorates glucose metabolism in a neonate streptozotocin-induced diabetic rat model through enhancement of cell proliferation (Irako et al. 2006) and protective effects against the pro-oxidant damage of diabetes (Turk et al. 2012). In a rat model in which ghrelin and GOAT genes are overexpressed in pancreatic β cells under the control of the rat insulin II promoter, the serum insulin levels, pancreatic insulin mRNA expression, and β-cell numbers in islets are increased (Bando et al. 2013), suggesting that acylated ghrelin can directly stimulate β-cell proliferation in vivo after islet injury even without its orexigenic or GH-stimulating activities. In other studies, both acylated and unacylated ghrelin prevent apoptosis in the β-cell lines HIT-T15 and INS-1E, as well as in human islets (Granata et al. 2007). Similarly, obestatin prevents β-cell apoptosis, preserves β-cell mass, and stimulates insulin secretion in vitro and in vivo in animals, in both normal and diabetic conditions (Egido et al. 2009; Granata et al. 2008, 2012). More recently, it has been observed that obestatin promotes the generation of islet-like cell clusters with increased insulin gene expression and C-peptide secretion (Baragli et al. 2013).
Mechanisms involved in a putative loop between autoimmunity machinery in the pancreas and the preproghrelin-derived peptides are still unexplored. However, administration of acylated ghrelin increased anti-insulin antibody (IA)-2β mRNA but not that of IA-2, another structurally related β-cell autoantigen in mouse brain, pancreas, and insulinoma cell lines (MIN6 and βTC3) (Doi et al. 2006). We observed that obestatin levels were negatively associated with C-peptide and IA levels at the onset of type 1 diabetes in a cohort of children with type 1 diabetes, advancing an interesting role in pancreas regulation in this condition (Prodam et al. 2014).
Follicular thyroid tissue of patients with Hashimoto’s thyroiditis produces ghrelin and obestatin at similar levels as in normal thyroid tissue (Bossowski et al. 2013; Karaoglu et al. 2009). Ghrelin also immunocolocalizes with calcitonin in normal C cells of different mammals (Utrilla et al. 2013), and C cells are also able to secrete ghrelin (Morillo-Bernal et al. 2011). GHS-R1a and GPR39, the discussed receptor of obestatin, have been detected more in thyroid tissues of patients with Graves’ disease than in those with non-toxic or toxic nodular goiter (Bossowski et al. 2013). Several data, often contrasting, reported ghrelin secretion in condition of hypo- and hyperthyroidism due to autoimmune and non-autoimmune diseases in humans. In the majority of the papers, ghrelin and obestatin are increased or unchanged in hypothyroidism, while they are lower or quite undetectable in hyperthyroidism. Thyroid substitution or tionamide treatment restores normal circulating levels (Bossowski et al. 2013). No correlations were observed between total ghrelin levels and antithyroid antibodies or human C reactive protein in children with untreated hyperthyroidism and subclinical hypothyroidism from Graves’ disease and Hashimoto’s thyroiditis, respectively (Sawicka et al. 2010). Conversely, total ghrelin levels were decreased in hypothyroid adults with high thyroid peroxidase antibody titer compared to hypothyroid patients with low thyroid peroxidase antibody with a negative correlation with age, thyroglobulin antibody, and thyroid peroxidase antibody titer (Altinova et al. 2006). In agreement with these findings, it is interesting to note that acylated ghrelin potentiates thyroid-stimulating hormone-induced expression of the thyroid tissue-specific genes thyroglobulin, thyroperoxidase, and sodium-iodine symporter in rat PC-Cl3 cells (Morillo-Bernal et al. 2011).
Autoimmune Gastrointestinal Diseases
Autoimmune Atrophic Gastritis
The majority of the studies on atrophic gastritis have been focused on the pathological condition due to a Helicobacter pylori infection. Only two studies have focused on atrophic gastritis due to the presence of parietal cell antibodies with contrasting results. The first one was conducted in a wide cohort of adults and reported lower circulating ghrelin levels in patients positive to parietal cell antibodies with a negative correlation between ghrelin and gastrin levels (Checchi et al. 2007). The second one was conducted in a small cohort of subjects affected by type 1 diabetes with or without atrophic gastritis. They observed that ghrelin secretion was lower in subjects with type 1 diabetes without a gastric disease, and conversely, ghrelin levels were similar to controls when an autoimmune atrophic gastritis is present at the same time, suggesting a compensatory ghrelin synthesis in neuroendocrine cell hyperplasia lesions of gastric mucosa (Alonso et al. 2007).
Celiac disease is one of the most prevalent and incident autoimmune diseases in the last years. It is characterized by malabsorption, weight loss, and increased energy expenditure. Higher total ghrelin circulating levels were recorded in children and adults with newly diagnosed celiac disease (Lanzini et al. 2006; Peracchi et al. 2003; Selimoglu et al. 2006) with a direct correlation with the degree of severity of intestinal mucosal lesions (Peracchi et al. 2003; Selimoglu et al. 2006). Patients recovered normal ghrelin levels after two years of gluten-free diet (Capristo et al. 2005; Lanzini et al. 2006). Accordingly, the mean number of ghrelin-positive cells/field in gastric and distal duodenal mucosa specimens was higher in celiac patients than controls and normalized after the gluten-free diet (Jarocka-Cyrta et al. 2010; Rocco et al. 2008). Many mechanisms have been advocated, including nutritional factors, functional responses to anorexia, malabsorption, and chronic inflammation. The role of higher ghrelin levels as a functional response to protect gut cells is supported by evidence that ghrelin administration protects against ischemia/reperfusion by reducing ulceration, tissue congestion, cellular infiltration, and vascular permeability (El Eter et al. 2007; Wu et al. 2009). In particular, ghrelin inhibits reactive oxygen species generation by human polymorphonuclear cells in a dose-dependent manner (El Eter et al. 2007). Interestingly, ghrelin administration is followed by an increased activity of stomach catalase, superoxide dismutase, glutathione reductase, and glutathione peroxidase in newborn streptozotocin-induced diabetic rats (Coskun et al. 2013).
Inflammatory Bowel Disease
Crohn’s disease is a chronic debilitating disease characterized by severe inflammation of the gastrointestinal tract. Th1 cells drive the inflammation of the colon. Patients with an active Crohn’s disease have higher total and acylated ghrelin levels than those in remission with similar obestatin levels and a higher ghrelin to obestatin ratio (Alexandridis et al. 2009; Karmiris et al. 2006). Similar data have been reported in ulcerative colitis and high levels of ghrelin correlate with the severity of disease and the activity markers (Ates et al. 2008; Karmiris et al. 2006). Infliximab increased circulating total ghrelin in a small group of patient with Crohn’s disease (Sung et al. 2009). Ghrelin mRNA levels in colonic mucosa of patients with inflammatory bowel diseases (Crohn’s disease and ulcerative colitis) were higher than in controls. The GHS-R mRNA level in colon and the percentage of peripheral blood T cells positive to this receptor in active Crohn’s disease, but not in ulcerative colitis, were also higher. However, lymphocyte reactivity to ghrelin was low in Crohn’s disease, suggesting that ghrelin might participate in its pathogenesis (Hosomi et al. 2008). Ghrelin administration abrogates the intestinal inflammatory response and restores mucosal immune tolerance in a mouse model of colitis down-regulating Th1-driven autoimmune response (Gonzalez-Rey et al. 2006). Both ghrelin and obestatin, although through different mechanisms, inhibit the inflammation and induce anti-inflammatory cytokines in a rat model of acute and chronic colitis caused by oral administration of dextran sulfate sodium (Pamukcu et al. 2013). In particular, in the acute state the administration of both peptides resulted in reduced lipid peroxidation and Th1 cells. In the chronic state, both ghrelin and obestatin decreased IL-1β, IFN-γ, TNF-α, but ghrelin increased TGF-β without affecting IL-10, while obestatin increased colonic levels of IL-10 and TGF-β (Pamukcu et al. 2013). Similar anti-inflammatory properties of ghrelin have been recorded in gut ischemia/reperfusion models, which require an intact vagus to maintain the beneficial effects of the hormone (Wu et al. 2009). These findings suggest that ghrelin attenuates inflammation and reduces injury through the modulation of the vagal cholinergic anti-inflammatory pathway. However, in contrast to the previous study, it has been demonstrated in a rodent colitis model, that ghrelin treatment enhanced clinical disease activity and promoted the infiltration of neutrophils and colonic IL-1β levels (De Smet et al. 2009). These results could depend on route and dosage of the peptide administration, as well as on the age of animals, and strongly suggest to be cautious in further studies, because without further clarifying the physiology of the ghrelin system, inadequate treatments could bring to contrasting or unsatisfactory results.
Therapeutic Applications of Ghrelin
In addition, in both human patients and experimental animal models, ghrelin administration ameliorates the cachectic state associated with several pathological conditions—such as chronic heart failure (Nagaya et al. 2004), chronic kidney disease (DeBoer et al. 2008), cancer (Argilés and Stemmler 2013; Neary et al. 2004; DeBoer et al. 2007), burn injuries (Balasubramaniam et al. 2006), and COPD (Nagaya et al. 2005)—likely through activity on both the immune system and the skeletal muscle (Reano et al. 2014).
Although ghrelin may undeniably inhibit cachexia through the GHS-R1a-mediated anti-inflammatory and orexigenic activities, some evidences prove that both acylated and unacylated ghrelin have a direct anti-atrophic activity in skeletal muscles: unacylated ghrelin, which does not bind GHS-R1a and does not activate the GH/IGF-1 axis, reduces burn-induced skeletal muscle proteolysis and TNF-α up-regulation in rats (Sheriff et al. 2012), counteracts muscle atrophy induced by either fasting or denervation, and, as a final point, both acylated and unacylated peptides impairs fasting-induced atrophy in Ghsr null mice (Porporato et al. 2013).
Ghrelin has a potent anti-inflammatory activity, and its ability to inhibit pro-inflammatory cytokines expression and release has been demonstrated by a huge number of studies, both in vitro and in vivo. The anti-inflammatory effects of the other products of the ghrelin gene—unacylated ghrelin and obestatin—have not been thoroughly investigated yet, although some studies attest their anti-inflammatory action on different tissues.
In both humans and animal models, ghrelin circulating concentrations increase in several inflammatory and autoimmune diseases that may be frequently associated with cachexia and anorexia. This increase may represent a compensatory mechanism of the organism in the attempt at re-establishing an optimal energetic balance, stimulating food intake, inhibiting muscle wasting, and fighting inflammation. Alternatively, as some experimental evidences in mice models of cachexia suggest, the increase in ghrelin levels in cachectic states may represent a hallmark of the establishment of a ghrelin resistance. On the other hand, in some diseases, it remains difficult to discern the primary cause for altered circulating ghrelin levels. It cannot be ruled out that in conditions of altered gut health like Crohn’s disease or atrophic gastritis, changed ghrelin levels are a feedback response to alterations in nutrient absorption, gastro-intestinal function, and caloric intake.
Several clinical trials were designed to treat cachectic patients with the administration of ghrelin, ghrelin analogs, and GHS-R1a agonists in order to stimulate food intake, increase body mass weight and functionality, reduce the inflammation, thus ameliorating the overall physical condition. The therapeutic potential of unacylated ghrelin and obestatin administration, on the contrary, is so far totally unexplored, although the available data on animal studies are encouraging and suggest that also these two peptides could represent a valuable strategy to treat inflammatory and autoimmune diseases. In addition, the use of unacylated ghrelin or obestatin could also represent an alternative to ghrelin treatment in conditions in which the activation of the GH-IGF-1 axis or the potential diabetogenic side effect of ghrelin could be more detrimental than beneficial for the patient.
The authors thank Omar Sabry for the critical reading of the manuscript and for the help with the figure.