Skip to main content

Towards improved carbohydrate utilization by finfish: physiological, metabolic, molecular and genetic limitations of poikilothermy


The overall objectives of the project were to understand the control mechanisms of the metabolic utilisation of dietary carbohydrates as affected by water temperature in three commercially important finfish species : rainbow trout, common carp and gilthead sea bream.

Firstly, the studies were aimed at the quantitative evaluation of the beneficial effects of different technological treatments to improve starch digestibility. Secondly, the project's aims were to understand the metabolic specificities between species indicating disorders of glucose homeostasis : prolonged postprandial hyperglycaemia, insulin secretion, receptivity, glucose phosphorylating capacity. The specific objectives were thus to a) assess the nutritional and endocrinological status or potential of fish fed two different sources and levels of carbohydrates using nutritional, biochemical and metabolic criteria and b) to evaluate the dietary or temperature influences on the activities of glucose phosphorylating enzymes and their induction at a molecular level.
I. Extrusion technology to improve starch gelatinisation and digestibility of starch

Different conditions of extrusion and flaking process were applied to whole-wheat grains and the digestibility of the obtained wheat products was measured in trout at 8°C and 18°C and in carp at 18°C and 25°C. The results show that when properly ground (< 500µ), the starch digestibility of whole wheat without any extrusion treatment is high in carp. However, for rainbow trout, a definite extrusion treatment is required to have the gelatinisation ratio above 98 % in order to obtain starch digestibility of 80 % (Figure 1).
II. Utilisation of dietary carbohydrates by rainbow trout and common carp

Four experimental diets having a common basis were formulated to contain one of two levels (20 and 35 %) of either extruded peas or extruded wheat. Apparent digestibility coefficients (ADC) of the experimental diets were measured at both temperatures. Two separate growth studies under experimental fish farm conditions were performed with juvenile rainbow trout and common carp, each reared at two different temperatures, in order to measure the efficiency of experimental diets containing the two different amounts of extruded wheat or peas. At the end of each trial, besides the determination of all nutritional parameters, tissues were sampled to study plasma glucose and insulin levels, hepatic hexokinase activity, expression of hexokinase mRNA and insulin receptors.

(For Figure 1 contact the Coordinator)

Figure 1. ADC of starch as affected by the gelatinization ratios in the rainbow trout and in the common carp at two rearing temperatures

Postprandial changes in plasma glucose and insulin

In fish fed a starch-free diet, the postprandial plasma glucose levels were low (below 1g/ltr) and relatively stable over the day. At 18°C, in trout fed starch rich diets, the peak values were observed between 6 to 8 hrs and slightly later in fish grown at 8°C. In trout grown at 18°C, the postprandial rise in plasma glucose levels was higher in those fed high levels of wheat starch than in all other groups. With regard to plasma insulin levels, no specific effect of the dietary treatments were found. Even in those fed a starch-free diet, the postprandial insulin pulse was high and this was not connected to any similar rise in plasma glucose values. This indicates that the increase in circulating levels of insulin is probably more due to an effect of dietary amino acid supply and utilisation by peripheral tissues rather than being directly related to the supply of dietary glucose.

(For Figure 2 contact the Coordinator)

Figure 2. Plasma glucose and insulin levels in trout fed either a starch-free diet (control or SG) or those fed low protein diets with two levels of extruded wheat starch (W 20 and W 35).

Hepatic glucose phosphorylating capacity in rainbow trout and common carp

Selected data on the hepatic HK and GK activities are reported in Figure 3. In rainbow trout grown at 18°C, liver HK activities were low but slightly affected by dietary starch levels. Activities of GK were very high and also affected by dietary starch availability. In trout fed the carbohydrate free diets, the glucose phosphorylating capacity seems very low. In common carp grown at 25°C, activities of GK were lower than in trout and the effects of dietary carbohydrates were less pronounced.

(For Figure 3 contact the Coordinator)

Figure 3. Hepatic HK and GK activities in rainbow trout and common carp as affected by dietary carbohydrate sources and levels.

Glucose oxidation and lipogenesis

Fish fed the experimental diet which led to the best growth performance and the highest lipid gain were used to evaluate the involvement of glucose in oxidative catabolism and in lipid synthesis. Postprandial changes in metabolic utilisation of glucose were studied by sampling fish 6, 12 and 18 hours after administration of a U-14C-glucose solution. Data revealed that glucose is actually used as an energy supplier especially at high temperature and that lipid synthesis from glucose is low.

Tissue receptivity to insulin and glucose transport

Plasma insulin levels increased after meal in all experimental groups and species. The number of receptors appeared to be however different between species possibly in relation with the natural dietary preferences of the three species. Initial studies show that in the rainbow trout, dietary carbohydrate level (20 or 40%) or source (wheat or pea) appeared to have no significant effect on postprandial insulin levels. In the muscle of trout, the number of insulin or IGF-I receptors did not show differences between groups (Fig. 4); in all cases, IGF-I binding was higher than insulin binding, and the Kd was lower for IGF-I receptors. In carp, the number of insulin receptors and specific binding were higher than in trout and were also positively correlated with the levels of carbohydrate in their diets. Insulin and IGF-I receptors showed tyrosine kinase activity, which was stimulated by the respective peptide and influenced by adaptation to dietary treatment.

Analysis of glucose transporters

The presence of a putative glucose transporter in fish that is homologous to the mammalian insulin-regulated glucose transporter GLUT4 was investigated using a molecular biological approach. Results show that the homology between mammalian and fish GLUT4 sequences is too low to enable the detection of fish GLUT4 mRNA with a mammalian GLUT4 cDNA probe by Northern blot analysis. Since GLUT4 cDNAs from non-mammalian species are not available, studies are under way to clone the trout GLUT4 cDNA by RT-PCR with degenerate oligonucleotide primers (DOP) based on conserved regions of GLUT4.

(For Figure 4 contact the Coordinator)

Figure 4. Number of insulin binding receptors from the skeletal muscle of trout and common carp.

III. Molecular cloning of the hexokinase (HK) and glucokinase (GK) genes in teleosts

By choosing adequate degenerate primers from a region corresponding to the glucose and ATP-binding sites in several species, and applying the DOP-PCR (Degenerated Oligonuclotide Primed Polymerase Chain Reaction) technique, DNA probes for GK from all the three species (trout, carp and sea bream) and for HK from the latter two were obtained. RNA hybridisation through Northern blot and specific RT-PCR for GKs from fish fed diets with and without carbohydrates were made. The expression of the rainbow trout GK gene as affected by the nutritional status was studied.

Figure 5. RNA hybridisation with the rainbow trout GK probe.
20 µg of total RNA from rainbow trout liver was submitted to electrophoresis on 1% agarose gel, transferred onto nylon membrane and hybridised with the probe. The membrane was hybridised with 16S common carp rRNA labelled and served as internal control.

(For Figure 5 contact the Coordinator)

As illustrated in Figure 5, the expression of the GK gene in rainbow trout livers differed significantly according to the dietary level of digestible starch in diets. There is a weak expression of the GK gene in livers of rainbow trout fed without carbohydrates (detected in PCR but not in Northern blot 6 hours after feeding) in contrary of the rainbow trout livers fed with carbohydrates. The GK mRNA expression was also found to be related to the level of hepatic GK activity measured in rainbow trout. These results strongly suggest that the GK gene expression in liver depends upon the dietary carbohydrate level in at least some teleosts.
The present interdisciplinary project was initiated to undertake growth studies with the three species fed two levels of two kinds of starch-rich ingredients (wheat, a cereal and peas, a pulse) and to analyse their effects on : nutrient / energy utilisation, phosphorylating capacity, postprandial glycaemia, insulin secretion and tissue receptivity and the levels of glucose transporters. A further objective was to do the molecular cloning of the hexokinase (HK) gene, whose activity is considered as the first limiting step in glucose phosphorylation. Selected data from the large amount of data generated so far (end of 2nd year) are shortly summarised below.


Station D'hydrobiologie
64310 Saint-pee-sur-nivelle

Participants (3)

Institut National de la Sante et de la Recherche Medicale
48, Bd Serurier 48 - Batiment Ecran +4 Et +5, Adr 12 - Paris-nord
75019 Paris
Universidad de Barcelona
645,Avda. Diagonal 645
08028 Barcelona
Universidade do Porto
Largo Prof. Abel Salazar 2
4000 Porto