Generation of transgenic mice overexpressing a ghrelin analog

After the discovery of ghrelin, we attempted to generate ghrelin gene transgenic (Tg) mice. These animals, however, produced only des-acyl ghrelin, which lacked the n-octanoyl modification at Ser(3) necessary to manifest ghrelin activity. Because the mechanism for acyl-modification of ghrelin had been unclear until the recent identification of GOAT (ghrelin O-acyltransferase), it had been difficult to generate Tg mice overexpressing ghrelin using standard procedures. Therefore, we planned to generate Tg mice overexpressing a ghrelin analog, which possessed ghrelin-like activity in the absence of acylation at Ser(3) and could be synthesized in vivo. As the replacement of Ser(3) of ghrelin with Trp(3) (Trp(3)-ghrelin) preserves a low level of ghrelin activity and Trp(3)-ghrelin can be synthesized in vivo, we generated mice overexpressing Trp(3)-ghrelin by using the hSAP (human serum-amyloid-P) promoter. Plasma Trp(3)-ghrelin concentrations in the Tg mice were approximately 85-fold higher than plasma ghrelin concentrations in non-Tg littermates. Because Trp(3)-ghrelin is approximately 1/10-1/20 less potent than ghrelin in vivo, plasma Trp(3)-ghrelin concentrations in Tg mice were calculated to have an activity approximately 6-fold greater than that of acylated ghrelin seen in non-Tg mice (85-fold x 1/10-1/20). Tg mice exhibited a normal growth and glucose metabolism in their early life stage. However, 1-yr-old Tg mice demonstrated impaired glucose tolerance and reduced insulin sensitivity. This model will be useful to evaluate the long-term effects of ghrelin or ghrelin analogs. In addition, this technique may be a useful method to generate gain-of-activity models for hormones that require posttranscriptional modifications.