Glucokinase (GK) plays a central role in glucose homeostasis in mammals. The absence of an inducible GK has been suggested to explain the poor utilization of dietary carbohydrates in rainbow trout. In this context, we analyzed GK expression in three fish species (rainbow trout, gilthead seabream, and common carp) known to differ in regard to their dietary carbohydrate tolerance. Fish were fed for 10 wk with either a diet containing a high level of digestible starch (>20%) or a diet totally deprived of starch. Our data demonstrate an induction of GK gene expression and GK activity by dietary carbohydrates in all three species. These studies strongly suggest that low dietary carbohydrate utilization in rainbow trout is not due to the absence of inducible hepatic GK as previously suggested. Interestingly, we also observed a significantly lower GK expression in common carp (a glucose-tolerant fish) than in rainbow trout and gilthead seabream, which are generally considered as glucose intolerant. These data suggest that other biochemical mechanisms are implicated in the inability of rainbow trout and gilthead seabream to control blood glucose closely.

Amino acids are considered to be regulators of metabolism in several species, and increasing importance has been accorded to the role of amino acids as signalling molecules regulating protein synthesis through the activation of the TOR transduction pathway. Using rainbow trout hepatocytes, we examined the ability of amino acids to regulate hepatic metabolism-related gene expression either alone or together with insulin, and the possible involvement of TOR. We demonstrated that amino acids alone regulate expression of several genes, including glucose-6-phosphatase, phosphoenolpyruvate carboxykinase, pyruvate kinase, 6-phospho-fructo-1-kinase and serine dehydratase, through an unknown molecular pathway that is independent of TOR activation. When insulin and amino acids were added together, a different pattern of regulation was observed that depended upon activation of the TOR pathway. This pattern included a dramatic up-regulation of lipogenic (fatty acid synthase, ATP-citrate lyase and sterol responsive element binding protein 1) and glycolytic (glucokinase, 6-phospho-fructo-1-kinase and pyruvate kinase) genes in a TOR-dependent manner. Regarding gluconeogenesis genes, only glucose-6-phosphatase was inhibited in a TOR-dependent manner by combination of insulin and amino acids and not by amino acids alone. This study is the first to demonstrate an important role of amino acids in combination with insulin in the molecular regulation of hepatic metabolism.

Most teleost fish are known to require high levels of dietary proteins. Such high-protein intake could have significant effects, particularly on insulin-regulated gene expression. We therefore analyzed the effects of an increase in the ratio of dietary carbohydrates/proteins on the refeeding activation of the Akt-target of rapamycin (TOR) signaling pathways in rainbow trout and the effects on the expression of several genes related to hepatic and muscle metabolism and known to be regulated by insulin, amino acids, and/or glucose. Fish were fed once one of three experimental diets containing high (H), medium (M), or low (L) protein (P) or carbohydrate (C) levels after 48 h of feed deprivation. Activation of the Akt/TOR signaling pathway by refeeding was severely impaired by decreasing the proteins-to-carbohydrates ratio. Similarly, postprandial regulation of several genes related to glucose (Glut4, glucose-6-phosphatase isoform 1), lipid (fatty acid synthase, ATP-citrate lyase, sterol responsive element binding protein, carnitine palmitoyltransferase 1, and 3-hydroxyacyl-CoA dehydrogenase), and amino acid metabolism (serine dehydratase and branched-chain α-keto acid dehydrogenase E2 subunit) only occurred when fish were fed the high-protein diet. On the other hand, diet composition had a low impact on the expression of genes related to muscle protein degradation. Interestingly, glucokinase was the only gene of those monitored whose expression was significantly upregulated by increased carbohydrate intake. In conclusion, this study demonstrated that macro-nutrient composition of the diet strongly affected the insulin/amino acids signaling pathway and expression pattern of genes related to metabolism.

Our objective was to analyse the hepatic transcriptomes of juvenile rainbow trout fed with a plantbased diet. We focused our analysis on the total replacement of fish meal (FM) and fish oil (FO) by a 100% plant-based diet (0% FM, 0% FO). We analysed the postprandial hepatic transcriptomes of rainbow trout fed the two diets 8 h after feeding. Six total hepatic RNAs from each dietary group were hybridised against a trout cDNA microarray (9K). After treatment of the data respecting the standard MIAME (Minimum Information About a Microarray Experiment) protocol, we found that 176 hepatic genes were differentially expressed between fish fed the two diets: 96 and 80 were over-expressed and under-expressed, respectively, in trout fed the plant-based diet. A large majority of differentially expressed genes were involved in metabolism (57%) and the others in cellular processes (21%) and transport (10%). Among the genes involved in metabolism (n = 86), 37% were associated with protein metabolism (proteolysis, amino acid catabolism), 21% with lipid metabolism (fatty acid biosynthesis, cholesterol biosynthesis), 30% with nucleic acid metabolism and 8% with glucose metabolism. Specifically, we found in rainbow trout fed the 100% plant diet an over-expression of genes involved in lipid biosynthesis (cholesterol metabolism and desaturation of polyunsaturated fatty acids) and an over-expression of a new metabolic actor, i.e., glycerol kinase which plays a key role at the interface of glucose-lipid metabolism. Overall, these data demonstrate that a number of intermediary metabolic effects occur in trout fed a totally plant-based diet.