After brief starvation, vertebrates maintain blood glucose by releasing fatty acids from adipose tissue. The fatty acids provide energy for gluconeogenesis in liver and are taken up by muscle, sparing glucose. After prolonged starvation, fat stores are depleted, yet blood glucose can be maintained at levels sufficient to preserve life. Using a new mouse model, we demonstrate that survival after prolonged starvation requires ghrelin, an octanoylated peptide hormone that stimulates growth hormone (GH) secretion. We studied wild-type mice and mice lacking ghrelin as a result of knockout of GOAT, the enzyme that attaches octanoate to ghrelin. Mice were fed 40% of their normal intake for 7 d. Fat stores in both lines of mice became depleted after 4 d. On day 7, mice were fasted for 23 h. In wild-type mice, ghrelin and GH rose massively, and blood sugar was maintained at 60 mg/dL. In Goat 2/2 mice, ghrelin was undetectable and GH failed to rise appropriately. Blood sugar declined to 20 mg/dL, and the animals were moribund. Infusion of ghrelin or GH prevented hypoglycemia. Our results support the following sequence: (1) Starvation lowers blood glucose; (2) glucose-sensing neurons respond by activating sympathetic neurons; (3) norepinephrine, released in the stomach, stimulates ghrelin secretion; (4) ghrelin releases GH, which maintains blood glucose. Thus, ghrelin lies at the center of a hormonal response that permits mice to survive an acute fast superimposed on chronic starvation.

The microsomal epoxide hydrolase 1 (EPHX1) Tyr113His and His139Arg polymorphisms have been reported to be associated with esophageal cancer (EC) risk, yet the results of these previous results have been inconsistent or controversial. The objective of this study was to explore whether the EPHX1 Tyr113His and His139Arg polymorphisms confer risk to EC. The relevant studies were identified through a search of PubMed, Excerpta Medica Database (Embase), Elsevier Science Direct, and Chinese Biomedical Literature Database until May 2013. The association between the EPHX1 Tyr113His and His139Arg polymorphisms and EC risk was pooled by odds ratios (ORs) together with their 95% confidence intervals (95%CIs). A total of eight case-control studies with 1163 EC patients and 1868 controls (seven studies for both Tyr113His and His139Arg polymorphisms, one study only for Tyr113His polymorphism) were eventually identified. We found no association between EPHX1 Tyr113His and His139Arg polymorphisms and EC risk in overall population (For Tyr113His: His vs. Tyr: OR = 1.05, 95%CI = 0.95-1.15, P = 0.379; His/His vs. Tyr/Tyr: OR = 1.04, 95%CI = 0.88-1.22, P = 0.208; His/Tyr vs. Tyr/Tyr: OR = 0.96, 95%CI = 0.80-1.15, P = 0.577; His/His vs. His/Tyr + Tyr/Tyr: OR = 1.10, 95%CI = 0.96-1.26, P = 0.164; His/His + His/Tyr vs. Tyr/Tyr: OR = 1.01, 95%CI = 0.90-1.12, P = 0.543. For His139Arg: Arg vs. His: OR = 1.04, 95%CI = 0.94-1.14, P = 0.465; Arg/Arg vs. His/His: OR = 1.06, 95%CI = 0.91-1.24, P = 0.470; Arg/His vs. His/His: OR = 1.03, 95%CI = 0.91-1.16, P = 0.673; Arg/Arg vs. Arg/His + His/His: OR = 1.04, 95%CI = 0.85-1.27, P = 0.708; Arg/Arg + Arg/His vs. His/His: OR = 1.02, 95%CI = 0.93-1.13, P = 0.617). In subgroup analysis based on ethnicity, significant association has been found in neither EPHX1 Tyr113His nor His139Arg polymorphism. The current meta-analysis suggests no evidence of association between the EPHX1 polymorphism and EC risk.