Unacylated ghrelin (UAG) accounts for approximately 80% of circulating ghrelin, but does not bind to or activate the receptor for acylated ghrelin (AG), the growth hormone secretagogue receptor (GHS-R) (1). Nevertheless, UAG exerts a distinct spectrum of central and peripheral actions, including the promotion of adipogenesis in bone marrow (2), but not in intra-abdominal white fat (WAT; 3). In the absence of an alternative receptor for this hormone, we investigated the potential interaction of UAG with GHS-R in bone marrow in more detail.As previously shown in rats (2), intra-bone marrow (ibm) infusion of AG and UAG (from an osmotic minimump connected to a tibial ibm catheter implanted under isoflurane anaesthesia) induced adipogenesis in wild-type (WT) mice, increasing the number of tibial marrow adipocytes from 22±3 cells/field (vehicle-infused; n=5) to 42±1 cells/field (AG; n=5; P<0.001) and 37±3 cells/field (UAG; n=4; P<0.05; all data are mean±SEM with statistical comparisons performed by 1-way ANOVA and Bonferroni post hoc test). Surprisingly, this effect was abolished in loxTB-GHS-R (GHS-R-null) mice (4), AG- and UAG-treated mice having 100 (n=5) and 88% (n=5) of the number of marrow adipocytes in vehicle-treated animals (n=5). Gas chromatography and mass spectrometry revealed that isolated rat tiibial marrow adipocytes contain a range of short-chain fatty acids, including octanoic acid, the substrate used for the activation of UAG. Fluorescence immunocytochemistry revealed that, unlike stomach (which expressed the activating enzyme, ghrelin O-acyl transferase (GOAT), but not the adipocyte-specific marker, PPARγ) and intra-abdominal WAT (which expressed PPARγ, but not GOAT), rat tibial marrow adipocytes co-expressed both PPARγ and GOAT. Immunogold electron microscopy revealed GOAT immunoreactivity in the membrane of the lipid trafficking vesicles and the plasma membrane of rat tibial marrow adipocytes. In addition, the adipogenic effect of a tibial ibm infusion of UAG (using the above method) was completely abolished in GOAT-null mice (5), UAG-treated tibiae (n=6) having 117% of the number or marrow adipocytes in vehicle-treated mice (n=5), compared to a 120% increase in adipocyte number in WT mice (n=5,6; P<0.05).Our data indicate that the adipogenic effect of UAG in bone marrow is dependent upon acylation by GOAT and subsequent activation of GHS-R. This suggests a novel endocrine mechanism, in which the target tissue is able to activate the incoming hormone, by the addition of the activating side-chain. The local expression of GOAT and GHS-R, together with the relative physico-chemical properties of AG and UAG may therefore determine the different activity spectra of these two hormones.