Several enteroendocrine cells have been identified in the stomach and shown to influence physiological functions with a predominant effect on gastric acid secretion, namely gastrin-producing cells (G cells), somatostatin-releasing cells (D cells, 5–10% of gastric oxyntic endocrine cells in rats, >20% in humans), enterochromaffin-like cells releasing histamine (ECL, 65% in rats, 30% in humans), and much less abundantly the serotonin-containing enterochromaffin (EC) cells (Rindi et al., 2004). In addition, a distinct cell type has been identified in the stomach that is distributed throughout the mucosa that was termed X/A-like cell in rats and P/D 1 cell in humans (Date et al., 2000; Mizutani et al., 2009). These cells were named X cells because of their unknown functions and in addition termed A-like cells due to their similarity with pancreatic A-cells (Rindi et al., 2004). They account for 20–30% of the oxyntic endocrine cells and therefore represent the second most abundant gastric endocrine cell type (Rindi et al., 2004). Distribution studies in the rat gastrointestinal tract indicate that the cell density (cells/mm 2) of X/A-like cells is 10- to 100-times greater in the gastric body than in the lower intestinal tract (Sakata et al., 2002). At the morphologic level, the X/A-like cells exist as closed-type cells without contact to the lumen and open-type cells with luminal contact. The open-type cells are more prominent in the ileum, cecum, and colon, where they account for more than 60% of ghrelin cells (Sakata et al., 2002). The identification of ghrelin in rat X/A-like and human P/D 1 cells (Rindi et al., 2002) as the only peripherally produced and centrally acting hormone known to increase food intake (Date et al., 2000) dramatically increased the interest in this endocrine cell type which is now commonly named ghrelin cell (Rindi et al., 2002).
Figure 1. Immunohistochemical photomicrograph of X/A-like cells in the rat gastric oxyntic mucosa of ad libitum fed male rats. Ghrelin-positive X/A-like cells (arrows) are evenly distributed throughout the entire length of the gastric oxyntic glands. The scale bar represents 50 μm.
Growing interest in ghrelin cells led to the discovery of additional peptide products derived from this cell type. These peptides are either derived from the same ghrelin gene including desacyl ghrelin and n-decanoyl ghrelin (Date et al., 2000; Hiejima et al., 2009) as well as obestatin (Zhang et al., 2005) or from a different gene, namely nucleobindin 2 (NUCB2)/nesfatin-1 (Stengel et al., 2009a).
Ghrelin was discovered in 1999 by Kojima and colleagues (reviewed in Kojima and Kangawa, 2011) and identified to be the endogenous ligand of the growth hormone (GH) secretagogue receptor 1a isoform (GHS-R1a; Kojima et al., 1999), which was later renamed ghrelin receptor (GRLN-R; Davenport et al., 2005). Ghrelin is a 28-amino acid peptide which has a unique n-octanoic acid residue on the serine-3 thereby increasing its lipophilicity (Kojima et al., 1999) and shown to be essential for binding to the GRLN-R (Kojima et al., 1999; Kojima and Kangawa, 2005). Structure-activity studies established that the first five N-terminal amino acids that include the hydrophobic residue are able to activate the receptor pointing toward the active core of ghrelin (Bednarek et al., 2000). Studies in mice ingesting different concentrations of medium-chain fatty acids (MCFA) or medium-chain triacylglycerols (MCT) established their direct use as a source for ghrelin acyl modification (Nishi et al., 2005). Without this post-translational modification desacyl ghrelin is obtained which does not bind to the GRLN-R. The gastric endocrine X/A-like cells are the major source of circulating desacyl and acyl ghrelin and the ratio of acyl and total (both acyl and desacyl) ghrelin in the circulation has been initially reported to be between 1:15 (Hosoda et al., 2000) and 1:55 (Raff, 2003). Recent improvements in blood processing resulted in a markedly higher acyl/total ghrelin ratio of 1:5 compared to 1:19 obtained after standard blood processing (EDTA blood on ice; Stengel et al., 2009b) indicating that although desacyl ghrelin represents the major form of circulating ghrelin, previous values were skewed by suboptimal blood processing conditions to preserve acyl ghrelin that is easily cleaved by a wide range of cellular protease and during protein extraction. Recently, another acylated form of ghrelin has been identified in humans and rodents, n-decanoyl ghrelin, which is also derived from X/A-like cells and circulates in considerable amounts in the mouse blood (Hiejima et al., 2009).
The enzyme catalyzing the acylation of ghrelin was unknown for a decade and recently identified in mice and humans as a member of the superfamily of membrane-bound O-acyltransferases (MBOATs), MBOAT4 that was renamed ghrelin-O-acyltransferase (GOAT; Gutierrez et al., 2008; Yang et al., 2008). GOAT is thought to octanoylate proghrelin before being transported to the Golgi apparatus where it is cleaved by prohormone convertase 1/3 (PC 1/3; Yang et al., 2008). Recently, GOAT protein was also identified in rodent circulation (Stengel et al., 2010d) leading to the possibility of an extracellular acylation of ghrelin. Both MCFAs C8 and C10 are substrates for GOAT-catalyzed acylation resulting in octanoyl and decanoyl ghrelin (Gutierrez et al., 2008). A current study reported the development of an antagonist of GOAT, a peptide-based bisubstrate analog GO-CoA-Tat, shown to be a useful tool in vitro and in vivo to assess the relevance of GOAT in body weight and glucose regulation (Barnett et al., 2010).
Alternative splicing and post-translational modification at a computer-based predicted cleavage site of proghrelin was reported to result in another biologically active peptide which was termed obestatin and assumed to have opposite effects to those of ghrelin (Zhang et al., 2005; Soares and Leite-Moreira, 2008). Obestatin immunoreactivity is also found in human gastric endocrine P/D 1 cells and localized in secretory granules (Gronberg et al., 2008; Tsolakis et al., 2009). Similarly, in rats obestatin fully colocalized with preproghrelin in intracellular dense core granules of gastric endocrine cells, whereas only partial (60%) colocalization of ghrelin and obestatin have been described giving rise to differential post-translational expression (Zhao et al., 2008).
NUCB2/nesfatin-1 was initially identified in the rat hypothalamus (Oh-I et al., 2006) but recently shown to be also expressed in the gastric oxyntic mucosa, prominently in gastric oxyntic endocrine cells (Stengel et al., 2009a). Colocalization of ghrelin and nesfatin-1 in rat gastric X/A-like cells was identified by immunofluorescence within different pools of vesicles indicative of a distinct subcellular distribution (Stengel et al., 2009a). Coexpression of these two peptides in X/A-like cells is also supported by the presence of PC 1/3 in this cell type (Yang et al., 2008) which is involved in the processing of both, ghrelin and nesfatin-1 (Yang et al., 2008; Shimizu et al., 2009).
Despite the fact that the functions of obestatin remain highly controversial (Goebel et al., 2008) and those of desacyl ghrelin (Chen et al., 2009) and nesfatin-1 (Garcia-Galiano et al., 2010) are just starting to be understood, all peptide products derived from this cell seem to be involved in the regulation of food intake with a stimulatory action of ghrelin and an inhibitory effect of desacyl ghrelin and nesfatin-1 (Stengel et al., 2010c).
Ghrelin-positive X/A-like cells represent by far the major source of circulating ghrelin (Ariyasu et al., 2001) as demonstrated by the sharp decrease of circulating ghrelin following gastrectomy (Jeon et al., 2004). In addition, lower amounts of ghrelin are produced in the intestine (Date et al., 2000), pancreas (Date et al., 2002b) and other peripheral organs including the kidney, liver, heart, testis, adipose tissue, and skin (Barreiro et al., 2002; Gnanapavan et al., 2002). Circulating ghrelin levels vary with metabolic status rising before and declining after a meal in various experimental animals and humans (Cummings et al., 2001; Tschop et al., 2001a). In addition, fasting increases gastric ghrelin mRNA expression in mice (Xu et al., 2009) and rats (Toshinai et al., 2001; Kim et al., 2003), whereas gastric ghrelin peptide content is decreased, indicative of increased synthesis and release of the peptide into the circulation by feeding (Toshinai et al., 2001
