The presence of protease inhibitors in different ingredients used in fish feeds may negatively affect their digestive utilisation. The effect of inhibitors present in several protein sources on the activity of digestive proteases of sea bream was assessed using ‘in vitro’ assays. Inhibition produced by extracts of plant proteins ranged from 25 to 50% of total activity, whereas that obtained using animal protein sources ranged from 1 to 20 %. The exception was ovoalbumin, which showed the highest measured value (62 %). A plot of the inhibition values obtained by changing the relative concentrations enzyme/inhibitor resulted in different curves for different protein sources. The use of sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) zymograms allowed visualisation of the aforementioned differences in inhibition. The importance of measuring protease inhibition in the preliminary evaluation of fish feed ingredients is discussed.

The effect of different parameters on short-term storage capacity of turbot ova was assessed over a 45-h period after ova collection for fertilization rates and over a 9-h period after ova collection for hatching rates. Increasing the volume of ova sampling from 0.5 to 2.5 mL, as well as adding an antibiotic-antimicotic solution or oxygen did not significantly change the storage capacity of ova. Regarding the hatching rates, a higher storage ability was recorded at 8 and 13 °C, compared to 3 °C. The mean composition of the ovarian fluid was determined (n = 57 spawns). Use of a diluent mimicking the ovarian fluid significantly decreased the storage ability as assessed by the fertilization rates but did not modify the hatching rates. Diluting ova in an artificial ovarian fluid deprived of calcium significantly decreased the fertilization and hatching rates during the storage period. Furthermore, addition or not of soybean trypsin inhibitor (Sigma T 9003) to the artificial ovarian fluid deprived of calcium did not significantly change the results. Storage capacity of control batches of ova was low: at 13 °C, without any diluent and when ova were fertilized 3 h after stripping, the hatching rate was lowered to 62.4 ± 29.4 % (mean ± SD) of the initial value.

The relationship between serum protein, immunoglobulin concentrations and protein molecular weight profile (PMWP) alterations of sea bass (131.3 ± 4.3 g) reared in sea water with sublethal concentrations of ammonia was studied over two periods totalling 116 days. During the exposure period (62 days) the first group (group 1) lived in sea water with 0.204 mg·L−1 unionized ammonia nitrogen equivalent to 12 % of the lethal concentration for 50 % of a population exposed for 96 h to ammonia (96-h LC50), whilst the second group (group 2) lived in sea water with 0.340 mg·L−1 UIA-N equivalent to 20 % of the 96-h LC50 of ammonia. Then, the two groups were left for a recovery period (54 days) in the same water as the control group (group 0). The determination of the total immunoglobulin (Ig) concentration was carried out by enzyme-linked immunosorbant assay (ELISA). Gel filtration columns were used for the serum PMWP. Serum Ig concentration of the exposed fish (group 2: 1.76 ± 0.43 mg·mL−1; group 1: 1.19 ± 0.33 mg·mL−1) was lower than the control group 13.39 ± 1.01 mg·mL−1) after 21 days of exposure period but this difference was reduced at the end of this period (day 62) and treated group Ig concentrations switched to higher than the control group after the recovery period (day 116, group 2: 9.75 ± 1.84 mg·mL−1; group 1: 7.50 ± 1.22 mg·mL−1 group 0: 6.38 ± 1.13 mg·mL−1). In fact, at the end of the experiment, the cumulative Ig production difference between fish exposed to ammonia and the control was less than 10%. Although a similar evolution of the Ig serum occurred with protein concentration, the serum protein concentration deficit of group 1 was restored at the end of the exposure period (group 2: 46.49 ± 2.34 mg·mL-1; group 0: 46.74 ± 1.97 mg·mL−1) and the cumulative production during the experiment was not significantly different between group 1 and group 0. However, this remained lower for group 2. During the exposure period, the PMWP of treated fish moved towards smaller molecular weight proteins. This alteration of the PMWP showed that the 0.2-kDa fraction increased, and another fraction appeared at the end of the exposure period. However, at the end of the recovery period, no difference was found between the PMWPs of control and treated fish. The alterations of the serum protein and Ig of fish reared, for a limited period (2 months), in water with sublethal concentrations of ammonia (20 % 96-h LC50 or less) should disappear completely after a few months in normal rearing conditions.

Results on changes in the total amino acids (protein bound + free) and the free amino acids (FAA) in relation to development, from egg (unfertilised and/or fertilised) to yolk-sac resorbed larva, before first feeding, in two Percichthyid fish, trout cod, Maccullochella macquariensis and Murray cod. Maccullochella peelii peelii, which lay demersal, adhesive eggs, are presented. Throughout development, the FAA accounted for only a small proportion (0.19% in fertilised eggs of both species) of the total amino acid pool. Nine essential amino acids (EAA) and eight non-essential amino acids (NEAA) were quantified in the amino acid pool at all stages of development. In both species, the total amino acid content decreased during the transformation (at 20 ± 1 °C) from newly hatched larva to yolk-sac resorbed larva. Overall, the changes in the TEAA and TNEAA reflected that of the amino acid pool. In trout cod, all but one EAA (lysine) and two NEAA (cysteine and glycine) decreased with ontogeny, from fertilised egg to yolk-sac resorbed larva. In Murray cod, however, the exceptions to the general decline were two NEAA (aspartic acid and glycine). In contrast, the FAA increased with development, the changes being reflected in both FEAA and FNEAA. Qualitatively, the predominant free amino acids in trout cod and Murray cod eggs were alanine, lysine, leucine and serine. Because the egg protein and the total amino acid contents declined with development, it is concluded that the rate of breakdown of yolk protein was higher than the anabolic and catabolic processes during embryogenesis. Data also suggest that in freshwater fish FAA are an unlikely primary energy substrate during embryogenesis.

A population of leaping grey mullet Liza saliens was studied in order to test whether growth rates decrease in summer owing to a reduction in food quality. A total of 330 juveniles of the 2-year age group were collected for measuring monthly changes in standard length, total weight, length growth rate and weight growth rate. Another 160 fish of the same age were caught monthly in February, April, August and November to measure food quality and calculate energy intake. Food quality changed as predicted. The organic matter content of the diet on a dry matter basis was 9.7% in April, 7.9% in August and 11.7% in November. The protein content of the diet on a dry matter basis was 9.0% in April, 6.4% in August and 9.3% in November. The carbohydrate content of the diet on a dry matter basis was 0.6% in April, 0.7% in August and 1.8% in November. The lipid content of the diet on a dry matter basis was 0.2% in April, 0.2% in August and 0.6% in November. The energy content of the diet was 2.63 kJ·g−1 in April, 2.21 kJ·g−1 in August and 3.13 kJ·g−1 in November. All the analysed mullet stomachs were empty in February, which suggests that they starve in winter. Daily rations increased when food quality decreased (4% of body weight in April, 6% in August and 1.4% in November). This allowed mullet to show high growth rates from June to August and to overcome a possible food limitation in the warm season. However, we do not know whether they would have grown even faster if food with a higher quality had been available.

Respiration rates of various geographical strains of Crassostrea gigas were compared to assess the respiratory expenditure as a physiological indicator of catabolism (cost of maintenance). Parental oysters, sampled in France (Marennes-Oléron), Japan (Hiroshima), Taiwan (Tunkang) and Spain (Cadix), were differentiated by both their geographic origin and by mitochondrial DNA markers, allowing the distinction between the two closely related taxa Crassostrea gigas and Crassostrea angulata. After reproduction of these parental oysters, respiration rates of spat of each strain, reared under common controlled conditions, were individually estimated at 20 °C by using a volumetric microrespirometer. Our results demonstrated that physiological variability existed among the Crassostrea gigas strains and is likely to be related to physiological differences between geographical regions and/or genetic adaptations. The French strain showed the highest rates (0.71 μL·h−1), while the Japanese, Taiwanese and the hybrid between Spanish and French strains consumed less, 0.53, 0.43 and 0.40 μL·h−1, respectively. Our results confirmed the discrepancies previously suggested between Crassostrea gigas and Crassostrea angulata. Comprehensive physiological assessments should be carried out over a wide temperature range to confirm our results and to further evaluate growth potential. Appropriate decision making based on these studies will help future shellfish management in shellfish rearing areas such as the overstocked Marennes-Oléron Bay.

The impact of Mediterranean mussels (Mytilus galloprovincialis Lamarck) on particle removal from the water column under different flow velocities was measured in the laboratory within a recirculating flume. At 20 °C, the filtration rates increased with current velocity up to 20−25 cm·s−1. Filtration rates remained the same at 26 °C. At high phytoplankton concentration (Cryptophyceae: 8 500−11 000 Cryptomonas sp. cells·mL−1), filtration rates were low (0.2−0.4 L·h−1 per standard individual of 1 g dry weight), whereas higher filtration rates (0.5−2.5 L·h−1 per standard individual) were observed with lower concentrations of the algae Prorocentrum minimum (Dinophyceae : 1 300−5 800 cells·mL−1). The filtration rates of M. galloprovincialis were affected by high phytoplankton loads, and exhibited a maximum related to current velocity. Our results suggest that mussels, feeding over a wide range of seston concentrations and flow velocities, show an adaptive behaviour in an area where environmental conditions are known to vary drastically.

Des paramètres de la reproduction d'une population isolée du tilapia Sarotherodon melanotheron ont été étudiés durant 2 années consécutives et comparés avec les données de la littérature obtenues sur des populations vivant dans son milieu naturel. Le sexe-ratio (année 1 : 1:2,69; année 2 : 1:1,96) en faveur des femelles varie de façon saisonnière. Les mâles atteignent la maturitésexuelle plus rapidement (année 1 : 129 mm LS ; année 2 : 126 mm LS) queles femelles (année 1 : 135 mm ; année 2 : 136 mm LS). Le cycle dematuration des gonades est continu. Ces résultats sont dans l'ensemblesemblables à ceux obtenus dans le milieu naturel de l'espèce. Cependant,une fécondité inférieure et un diamètre ovocytaire plus grandsont notés dans le lac d'Ayamé. Le facteur de condition moyen despopulations du lac, intermédiaire entre des valeurs observées dans lesmilieux lagunaires, indique une bonne adaptation de l'espèce aux conditionsd'eau douce du lac.

The relationship of size to MEP-2* genotype was investigated in 0+ juvenile Atlantic salmon (Salmo salar) in the River North Esk, Scotland. Significant temperature-associated MEP-2* heterogeneity was observed among tributaries and, in the two tributaries studied, size and MEP-2* genotype were significantly associated. No size association, or heterogeneity among tributaries, was found for the other allozyme polymorphisms screened. The results suggest that the association between size and MEP-2* variation, noted by others, is likely to be widespread in wild populations, and potentially linked to habitat temperature.