Ghrelin is an acylated peptide with growth-hormonereleasing activity (1 ). It was first isolated from rat and human stomach during the search for an endogenous ligand to the "orphan" G-protein-coupled receptor, growth hormone secretagogue receptor (2 ). The peptide contains 28 amino acids, and n-octanoylation of the Ser-3 hydroxyl group is necessary for biological activity. Most studies have focused on the somatotropic and orexigenic roles of ghrelin; therefore, little is known about the kinetics of this peptide. Because the ester bond is both chemically and enzymatically unstable, elimination of the octanoyl modification of ghrelin can occur during storage, handling, and/or dissolution in culture medium (3 ). Because of increased interest in ghrelin measurements, a standardized method of sample collection is required.In the present study, which focused on the active form of ghrelin, we investigated the effects of anticoagulants and storage conditions on ghrelin stability. To distinguish the active form of ghrelin, we established two ghrelinspecific RIAs; N-RIA recognizes the N-terminal, octanoylmodified portion of the peptide, whereas C-RIA recognizes the C-terminal portion. Thus, the value determined by N-RIA specifically measures active ghrelin, whereas the value determined by C-RIA gives the total ghrelin immunoreactivity, including both active and desacyl ghrelin (4 -6 ). The minimum detectable quantities in the N-and C-RIAs were 5.0 and 50 pmol/L, respectively. The respective intra-and interassay CV were 3% and 6% for the N-RIA and 6% and 9% for the C-RIA (n ϭ 8 assays). Data are reported as the mean (SD). Comparisons of the time course of ghrelin concentrations between subgroups were made by two-way ANOVA for repeated measures, followed by the Scheffé test. P Ͻ0.05 was considered statistically significant.All blood samples were taken from three healthy male volunteers who gave written informed consent. Blood was taken from the forearm vein and immediately divided into tubes for serum and plasma preparation using (a) disodium EDTA (1 g/L) with aprotinin (500 000 kIU/ L), (b) disodium EDTA alone, (c) heparin sodium, or (d) no anticoagulant. Synthetic human ghrelin was added to each blood sample at a final concentration of 40 g/L; each sample was then sequentially divided into two aliquots for incubation at either 4 or 37°C. After incubation for 0, 30, and 60 min, blood samples were centrifuged, diluted 1:200 in RIA buffer, and subjected to ghrelin-specific RIAs. A comparison of the effects of different anticoagulants on the detected ghrelin concentrations is shown in Table 1A. Although the serum and three different plasma samples tested gave comparable results for total ghrelin by C-RIA, the N-RIA gave ghrelin concentrations that were significantly decreased at 37°C. When the ghrelin was measured by N-RIA, serum samples were highly affected by such treatment; samples stored for 60 min at 37°C lost ϳ35% of the ghrelin compared with the basal values at 0 min (P Ͻ0.05). The ghrelin concentrations in samples co...