Evaluation of Feed and Feeding Regime on Growth Performance, Flesh Quality and Fecal Viscosity of Atlantic Salmon (Salmo salar L.) in Recirculating Aquaculture Systems

SUN Guoxiang, LIU Ying, LI Yong LI Xian and WANG Shunkui

1) Institute of Oceanology, Chinese Academy of Sciences, Qingdao 266071, P. R. China

2) University of Chinese Academy of Sciences, Beijing 100049, P. R. China

3) Shandong Oriental Ocean Sci-Tech Co., Ltd, Yantai 264003, P. R. China

Evaluation of Feed and Feeding Regime on Growth Performance, Flesh Quality and Fecal Viscosity of Atlantic Salmon (Salmo salar L.) in Recirculating Aquaculture Systems

SUN Guoxiang1),2), LIU Ying1),*, LI Yong1), LI Xian1), and WANG Shunkui3)

1) Institute of Oceanology, Chinese Academy of Sciences, Qingdao 266071, P. R. China

2) University of Chinese Academy of Sciences, Beijing 100049, P. R. China

3) Shandong Oriental Ocean Sci-Tech Co., Ltd, Yantai 264003, P. R. China

? Ocean University of China, Science Press and Spring-Verlag Berlin Heidelberg 2015

The effects of different feeds and feeding regimes on growth performance, flesh quality and fecal viscosity of Atlantic salmon (Salmo salar L.) in recirculating aquaculture systems (RAS) were investigated. Fish (initial body weight of 1677 g ± 157 g) were fed with four commercial feeds (Nosan salmon-NS, Aller gold-AG, Skretting salmon-SS and Han ye-HY) in two feeding regimes (80% and 100% satiation) for 78 d. The results showed that salmon specific growth ratio (SGR) and weight gain ratio (WGR) were significantly affected by feed type and feeding regime (P ＜ 0.05). Feed conversion ratio (FCR) varied between 0.93 and 3.40, which was significantly affected by feed type (P ＜ 0.05), and slightly improved with increased satiation degree. The activities of digestive enzymes including protease, lipase and amylase were also significantly affected by feed type and feeding regime (P ＜ 0.05), increasing with satiation degree. Flesh qualities for vitamin E, hydroxyproline (HYP), liquid loss and muscle pH among all groups showed significant differences (P ＜ 0.05), ranging from 26.67 to 29.67, while no obvious difference was found in flesh color. Fecal viscosity for different treatments showed no significant difference, though improvement was found in 100% satiation group. From present experiment, it was concluded that both feed type and feeding regime can affect the important quality attributes of Atlantic salmon.

Atlantic salmon; Salmo salar L.; recirculating aquaculture systems; RAS; feed; feeding; growth; flesh quality; fecal viscosity

1 Introduction

Currently, interests in the culture of Norwegian Atlantic salmon (Salmo salar L.) have increased greatly all over the world in last decade as salmon possesses a number of attributes desirable for commercially cultured fish for high flesh quality (FAO, 2011). Fish germ cells were hatched on freshwater and then cultured in freshwater or seawater after smolting, either in ponds or cages (Burr et al., 2012). Although the vast majority of farmed salmon are currently produced by net-cage operations, adoption of recirculating aquaculture systems (RAS) for salmon production can mitigate some issues associated with the development of aquaculture in Europe and other countries (Martins et al., 2010). These issues include water and land restrictions by limited supply or by regulations put in place to conversion for other purposes. Additionally, environmental sustainability requirements forland-based systems have become stronger in recent years, increasing the regulations for effluent discharge and the need for advanced cleaning systems to treat aquaculture effluents (Unger and Brinker, 2013). However, as a feasible alternative, RAS must be effectively and economically viable, considering the combined costs of feed and feeding operations can reach up to 60% of total operational expenses (Tucker, 1998). Feeding methods can result in 1%–30% of the feed applied to the system to pass by the fish uneaten depending on the feeding strategy employed (Summerfelt, 1998). Therefore, development of cost-effective feeds and feeding strategies is of great importance for future development of salmon aquaculture, especially for RAS technology.

As one of the vast cultured species, scientific literatures on salmon nutrition have developed rapidly in recent years. Researches evaluating ingredient digestibility (Pratoomyot et al., 2010), essential amino acid requirements (Torstensen et al., 2008), optimal crude protein, crude lipid inclusion (Crampton et al., 2010), and substitutes for fishmeal and fish oil in the formulations to re-duce cost and environmental impacts (Gatlin et al., 2007; Miller et al., 2008; Tacon and Metian, 2008; Naylor et al., 2009; Turchini et al., 2009) have been covered in recent years. These characteristics, coupled with successful expansion and growth of salmon aquaculture all over the world, have served as stimulants for salmon farming development in traditional culture systems (Bendiksen et al., 2011), especially for marine cage culture. Nonetheless, no commercial feed specifically for Atlantic salmon in RAS are currently available in China. Therefore, farmers have to use the diets developed for other salmonids. Knowledge for salmon nutrition is still limited until sufficient nutritional information is established under different culture conditions, including RAS.

The present study was conducted to compare the effect of four available commercial diets with different protein and lipid contents on the growth performance, flesh quality and fecal viscosity of salmon reared in RAS. The effect of feeding regime on the above indicators was also evaluated. The results will contribute valuable information to further manipulation of commercial feed in RAS.

2 Materials and Methods

2.1 Fish and Feed

The experimental Atlantic salmon (Salmo salar L.) were obtained from Oriental Ocean Sci-Tech Co., Ltd (Yantai, China). On March 21 and 22 2012, fish (initial body weight of 1677 g ± 157 g) were transferred into experimental RASs at an initial stocking density of (20.05 ± 0.95) kg m-3, equivalent to 1050 fish per tank. After stocking, fish were handfed to four diets (Table 1): Nosan salmon (NS, 480 g kg-1crude protein, 160 g kg-1crude fat), Aller gold (AG, 400 g kg-1crude protein, 280 g kg-1crude fat), Skretting salmon (SS, 480 g kg-1crude protein, 180 g kg-1crude fat) and Han ye (HY, 440 g kg-1crude protein, 220 g kg-1crude lipid) twice a day at 8:00 and 16:00. Each of the four feed groups was separated into two experimental groups, with one fed to satiation and the other fed restrictively to 80 percents of the satiation groups (80% feed amount of the satiation groups). Thus, eight experimental groups (two tanks each diet) were created and named according to the feed groups and the following feeding regimes.

Table1 Proximate composition of four commercial diets fed to salmon during the experimental period

2.2 Experimental Systems and Culture Conditions

The present study was conducted in replicated RASs located at Oriental Ocean Sci-Tech Co. Ltd (Yantai, China) from March 23 2012 to June 8 2012. Each RAS consists of four 88 m3round culture tanks and supplementary facilities, i.e., mechanical filtration, foam separator, UV disinfection and moving bed biofilter (Fig.1). Water was supplied to each tank at a rate of approximately 10 m3h-1with liquid oxygen of 8.0–10.0 mg L-1. During the experiment period, the photoperiod was maintained at 24 L: 0D and daily renewal water rate was 100%. Water quality variables were temperature 15.70℃ ± 0.40℃, pH 6.50–7.50, salinity 24–26 and dissolved oxygen ＞ 8.00 mg L-1.

Fig.1 Recirculating aquaculture systems used in the present study (RAS, N=8). Fish sampling positions for final body weight measurement are marked with +.

2.3 Sampling Procedure

At the start of the trial, fish were starved for 24 h and individual weights were recorded to nearest 1.0 g to determine initial body weight. At the end of the trial, fish were starved for 24 h, and 120 fish in four places (Fig.1) each tank were sampled and weighed to nearest 1.0 g to determine production characteristics. Feces collection started from 3 April twice a week. Feces from swirl separator were recovered 3 h after feeding and were stored at -80℃ until analysis. At the final sampling, 30 fish of each tank was randomly collected and exposed to anesthetic of MS-222 for 1 min. Plasma for enzyme analysis was collected from the caudal vein into sterilization centrifuge tube (heparin sodium as anticoagulant) and kept on ice. Then serum was obtained through centrifugation at 5000 g at 4℃ for 10 min, and samples were stored at -80℃ until analysis. Fish was killed with a sharp blow to head, and tissues (liver, foregut and stomach) were separated, rinsed with 4℃ deionized water, and then stored at -80℃ for further analysis. Muscle pH (30 fish per group) was determined within 40 min post-mortem by inserting a pH electrode (portable meat pH metre, HI99163, Hanna Instruments Ltd, UK) into the Flesh Quality Cut, (FQC); dorsal part of the fillet posterior to the head (Fig.2). The Scottish Quality Cut (SQC) and the Norwegian Quality Cut (NQC) were obtained from each salmon fillet (Fig.2) and used for further analyses of flesh characteristics such as connective tissue (hydroxyproline, HYP), colour, vitamin E content, and liquid loss.

Feces sampling for rheological measurements were carried out between June 1 and 6 according to Brinker(2009). Five fish were sampled daily from each tank from 8:40 to 10:00. Individual fish were selected randomly, anaesthetized with MS-222 (dose: 40 mg L-1, exposure time: 1 min) and then killed by a sharp blow to the head. Mucus fecal pellet was collected from intestinal dissection. Feces were placed in aluminum dishes, hermetically sealed with a plastic film to prevent dehydration and cooled to 4 ℃ to slow down microbial degrad ation. All measurements were performed within 8 h of dissection.

Fig.2 Sample sites for determination of flesh characteristics with the Flesh Quality Cut (FQC), the Scottish Quality Cut (SQC) the Norwegian Quality Cut (NQC) according to Johnsen et al. (2011). Muscle pH in right side of FQC, flesh colour in right fillet of NQC, liquid loss and HYP in left fillet of SQC, and vitamin E in the left fillet of NQC.

2.4 Feed Analysis

Analysis of feed was made following the procedures of AOAC (2003). Total nitrogen content of feed was analyzed using Dumas method (Ebling, 1968) with a KDY-08A nitrogen/protein determinator (Shanghai Ruizheng Instrument Co., Ltd., China), and crude protein content was calculated (Nitrogen×6.25). The content of crude lipid was measured by ether extraction (34.6 BP). Gross energy content of fish, feces and diet was measured by oxygen bomb calorimeter (Phillipson microbomb calorimeter, Gentry Instruments, Aiken, SC, USA).

Feces and feed acid insoluble ash (AIA) was measured according to Goddard and Mclean (2001).

2.5 Calculation of Growth and Apparent Digestibility

The growth indicators were calculated by the following equations:

where SGR is specific growth rate, HIS is Hepatosomatic index, Wfis final weight, and Wiis initial weight.

The dry matter apparent digestibility coefficient was calculated by the following formula (Windell et al., 1978):

where ADC is apparent digestibility coefficient.

2.6 Activity of Digestive Enzymes

2.6.1 Protease activity in stomach and intestinal

Protease activity of stomach crude extract was determined according to (Kunitz, 1947) using casein as the substrate. A mixture of 0.l mL crude extract, 0.5 mL phosphate buffer (100 mmol L-1, pH 7.5) and 2 mL casein (1% in phosphate 50 mmol L-1, pH 7.5) was incubated at 37℃ for 20 min. The reaction was stopped by adding 1 mL of TCA (tri-chloro acetic acid, 30%) and the mixture was clarified through centrifugation (5000 r min-1for 15 min). Absorbance was measured at 440 nm (Spectrophotometers 2000, Bausch and Lomb, Rochester, NY, USA). An additional negative control was made by replacing crude extract with phosphate buffer. A standard curve of absorbance at 440 nm was established by using tyrosine as standard. One unit of protease activity was defined as the quantity of micromoles of tyrosine released per minute per gram of protein at 37℃.

2.6.2 Lipase activity in intestinal

Lipase activity was assayed with a kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China). Trireactive glyceride and water became emulsoid, which can absorb and scatter the incident ray as it passed. The turbidity decreases when the emulsoid split as the trireactive glyceride is hydrolyzed with the help of lipase and the speed is proportional to the activity of lipase in the specimen. One unit of lipase activity was defined as the micromole number of substrate released per minute per gram of lipid at 37℃. All the above three enzyme tests were made in three replications.

2.6.3 Amylase activity in intestinal

Amylase activity was determined according to Rick and Stegbauer (1984) with 1% starch solution as substrate in 50 mmol L-1phosphate buffer with pH 7.5. After adding 0.1 mL crude extract into 1 mL substrate, the mixture was incubated at 37℃ for 5 min. The activity was determined by measuring the production of maltose resulting from starch hydrolysis. The maltose production was estimated by reading the colour intensity at 550 nm. A negative control was made by replacing crude extract with phosphate buffer. A standard curve of absorbance at 550 nm is established by using standard maltose solution. One unit of amylase activity was defined as the number of ten-micromole of maltose released per minute per gram of protein at 37℃.

2.7 Flesh Quality Measurements

2.7.1 Flesh color evaluation

The color of fast muscle was measured at six positionson the right NQC fillet according to previous study (Johnsen et al., 2011) using the DSM SalmoFan? Lineal (DSM, Switzerland). The SalmoFan color was evaluated under standardized light conditions (Ra N90, color temperature N5000 K) according to the Norwegian Standard (NS-9402E, 1994).

2.7.2 Vitamin E measurement

The tissue sample (1.5 g) was weighed to 0.0001 g, placed in conical flask, homogenized in 30 mL absolute ethylalcohol, 10 mL of 10 g mL-1ascorbic ethylalcohol and 10 mL of 50 g L-1sodium hydroxide solution, and then incubated at 70℃ for 30 min. The flasks were flushed with 60 mL ultrapure water, and the flushing fluid was transferred to separating funnel. After adding 50 mL petroleum ether, the funnels were returned to the water bath for 20 min with shaking 5 min, and the layers started to separate. Removing the aqueous phase and extracting it for 2 times, the three petroleum ether layers were mixed and flushed with 150 mL petroleum ether saturated water. Then the mixer was dehydrated with sodium sulphate anhydrous, removed to brown glass flask volumetric, added 100 mL BHT and stored at -20℃ until analysis. Measurement of vitamin E was conducted using a 150 mm × 4.6 mm reverse phase column (Luna 5μ C18; Phenomenex, Macclesfield, UK) using a mobile phase of methanol/ultrapure water (98:2) with a flow rate 1 mL min-1. The chromatographic system was equipped with a Waters Model 510 pump, and vitamin E isomers were detected at 292 nm using a Waters 490 Emulti wave length UV/VIS detector (Waters Ltd., Watford, England). Sample concentrations were calculated using standard tocopherol (Sigma, T3634).

2.7.3 Hydroxyproline assay

Alkaline-soluble (a-s) hydroxyproline (HYP) was analyzed from fast muscle in duplicates according to an established method (Hagen et al., 2007). The samples of fast muscle were obtained from standard blocks as described by Johnsen et al. (2011). They were homogenized in a food processor and stored at -40℃ until analysis.

2.7.4 Liquid loss and muscle pH assay

The liquid loss was determined gravimetrically according to a net test described by Ofstad et al. (1993). Homogenized duplicate samples (2×15 g) of post-rigor fast muscle, obtained from the muscle blocks (Fig.2), were centrifuged at 210 g at 10℃ for 15 min. The liquid loss was calculated as the percentage of liquid mass released during centrifugation.

Muscle pH (10 fish per group) was determined within 40 min post-mortem by inserting a pH electrode (portable meat pH meter, HI99163, Hanna Instruments Ltd, UK) into the Flesh Quality Cut (FQC)-dorsal part of the fillet posterior to the head.

2.8 Fecal Viscosity

Fecal viscosity was carried out on merged samples with a volume of 5 cm3feces. The fecal samples were transferred to a Paar Physica UDS200 rheometer. The shear stress factor was 2.037 and shear rate factor was 2.617. For the time sweep a deformation with an amplitude of c = 30% at a frequency of 1 Hz was used. The duration of measurement was set at a logarithmic scale using 21 intervals (between 0.1 and 100 s log) with continuous frequencies (0.10–100 Hz). In each run the deformation amplitude was 10% and the duration of measurement was 30 s. In the sample compartment, the temperature was set to 4℃to slow down the microbial action.

2.9 Data Analysis

The statistical analyses were performed using SPSS version 18.0. Data were presented as mean±SEM and analyzed by two-way ANOVA. The effects of feed type and feeding regime (Sat and Res) on digestive enzyme activity and flesh quality were tested by analysis of covariance (ANCOVA), with body mass as a covariate. The rheological data of the time sweep measurements were analyzed using repeated measurement ANOVA with the variable measuring point as a random block factor. The first ten points from the rheological data were excluded from the analysis in order to avoid anomalies resulting from the possible presence of undetected air bubbles within the samples. The rheological data for the frequency sweep measurements were analyzed using nested analysis of covariance (ANCOVA) with the variable tank as a random factor. The dependent variable viscosity and the independent variable frequency were log10-transformed to meet the assumptions of the model. Significance was accepted at P-value of 0.05.

3 Results

3.1 Growth Performance

High survival rate was observed during the 78 days feeding period, representing mortalities from 0.48% to 5.36% for all groups. In groups of feed type, significant differences (P ＜ 0.05) were still found among treatments (Table 2). For the two feeding treatments, restricted feeding (80%) have lower mortality than satiation groups (100%) by 53.70%–115.30%. Weight gain rates (WGR) ranged from 16.47% ± 3.73% to 79.31% ± 6.91%, which were significantly affected by feed type and feeding regimes. Similar to WGR, specific growth rates (SGR) in HY groups (0.70% ± 0.06% to 0.77% ± 0.01% BW d-1) were significantly higher than others (0.20% ± 0.04% to 0.55% ± 0.04% BW d-1), and satiation groups had higher SGR than restrictively feeding groups. Feed conversion rate (FCR) varied from 0.93 to 3.40. There was no significant difference for FCR among groups of AG, SS, HY and 100% NS, except for 80% NS, which showed lower FCR value than former treatments. Fish fed to satiation had better FCR than restricted groups. No significant differences in HSI were found and restricted groups had lower value than satiation. The dry matter ADC in all treatments were affected only by feed type and ranged from 69.19% ± 4.22% to 87.51% ± 0.13%.

Table 2 Growth performance and dry matter apparent digestibility coefficients (DMADC) of salmon fed with different feeds and satiations

3.2 Digestive Enzyme Activity

Digestive enzyme activities of protease in stomach and intestinal, lipase and amylase in gut were significantly affected (P ＜ 0.05) by feed type and satiation degree (Table 3). Protease activities in stomach and gut increased with satiation, with value of 0.59 ± 0.06 to 2.03 ± 0.11 U mg-1protein and 1.17 ± 0.04 to 2.96 ± 0.06 U mg-1protein. On the other hand, lipase and amylase in gut also increased with increased satiation, with ranges of 10.66 ± 0.67 to 41.92 ± 0.47 U g-1protein, and 82.54 ± 6.48 to 181.86 ± 3.54 U g-1protein, respectively.

Table 3 Digestive enzyme activity of stomach protease (Sprotease), intestinal protease (Iprotease), intestinal lipase and intestinal amylase of salmon fed with different feeds and satiations

3.3 Flesh Quality Characteristics

Flesh quality characteristics for different treatments in the present study were shown in Table 4. Flesh color was not affected by feed type or feeding satiation degree, with the value of 26.67 ± 0.44 to 29.33 ± 0.60. In the final sampling, HY groups were slightly higher than NS groups by nearly 3 units and flesh color increased with satiation degree. Vitamin E concentrations in wet weight and protein were significantly affected by feed type and satiation degree. The two forms of vitamin E increased with satiation degree, with the ranges of 0.21 ± 0.04 to 0.29 ± 0.02 mg kg-1and 6.43 ± 0.43 to 11.85 ± 0.49 mg kg-1. HYP concentration of fast muscle was significantly affected by feed type and found to be 63% higher in NS group than in AG group. On the other hand, satiation degree showed negative correlation with HYP. Liquid loss and muscle pH were both affected by feed type and feeding regime. For liquid loss, NS group was found to be nearly 35% higher than HY group. In addition, liquid loss increased with satiation degree and NS100% group showed the highest loss, while HY80% group had the lowest loss. Similar effects of different satiation degrees were found on muscle pH, with the value of 6.19 ± 0.02 to 6.36 ± 0.00. However, SS group had higher pH than AG group.

3.4 Fecal Viscosity

The rheological properties of feces for all treatment groups were shown in Table 5. Feed type and feeding satiation had no significant effect on fecal viscosity and there was no significant interaction between them. However, improvement of 100% satiation compared to 80% was still found during the feeding trial, with a range from 93.20% to 441.11%. For evaluating shear thinning, the relationship between frequency sweep and fecal viscosity was shown in Fig.3. These measurements were used to detect differences in the negative slopes of the linear fits. A lower negative slope would indicate greater structural resistance to mechanical stress. The statistical analysis revealed no significant difference in the slopes. However, better rheological properties of feces can still be seen in groups of HY with 80% satiation.

Table 4 Flesh quality, vitamin E, HYP, liquid loss and muscle pH of salmon fed four diets and two satiation degrees

Table 5 Viscosity of feces of salmon fed four diets and two satiation degrees

Fig.3 Frequency sweep data of the rheological measurements including regression lines. Data points of treatments are displaced horizontally by 6%.

4 Discussion

Feed intake is believed to be regulated by diet energy and feeding regime (level and frequency) to meet fish energy requirement under various culture conditions (Fauré and Labbé, 2001). Previous studies have demonstrated the lipid (Johansen et al., 2001, 2002, 2003) and protein (Bendiksen et al., 2003; Azevedo et al., 2004a; Solberg, 2004) regulation of feed intake of salmon. Recent development of salmon feed has led to the use of energy dense diets containing low protein and high lipid (Solberg, 2004). Low-protein and high-lipid diets are of potential benefits as they require less fish meal consumption. Previous researches have stated that low protein/ high lipid diets had no negative effect on fish growth and feed utilization (Hillestad et al., 1998; Bendiksen et al., 2003; Azevedo et al., 2004b). Although FCR in the present study seemed to be slightly improved with reduced protein/lipid ratio, this was not statistically significant except for groups of NS80%. Similar tendencies have been found in fish mortality and HSI. However, reduced protein/lipid ratios have significant effects on fish WGR and SGR. AG and HY diets with low protein/lipid ratios (1.43 and 2.00) resulted with better growth WGR and SGR than others. This can be explained by the sparing of protein by increased dietary lipid (Einen and Roem, 1997). ADC of dry matter in this study was also significantlydifferent among all groups and AG was with the higher value (86.90% ± 1.65% to 87.51% ± 0.13%). This might be due to the different feed materials and formulations. Restrictively fed groups were provided 80% of the ratio compared to the 100% satiation groups. Thus, growth performance was reduced in the 80% satiation groups and this result was in line with other reports (Jobling, 1994; Guillaume et al., 2001).

Feed type and satiation degree affected protease and lipase activities in stomach and intestinal sections in similar patterns–a decreased with reduced protein/lipid level in diet and feeding level, whilst amylase activity showed no significant difference for all groups. This result was in line with the results on flounder (Paralichthys olivaceus) (Li et al., 2005), jade perch (Scortum bacoo) (Shao et al., 2004) and shrimp (L. vannamei) (Le Moullac et al., 1997). The less feeding level (80% satiation group) with lower protease, lipase and amylase activities in stomach and intestinal is possibly due to nutrients quantity reduction, which can lead to protein degradation (Houlihan et al., 1988) as well as decreased synthesis (McMillan and Houlihan, 1989).

Flesh color (astaxanthin concentration) has been demonstrated to be related with diet energy and feeding regime (Bjerkeng et al., 1997; Jensen et al., 1998). In the present study, statistical significance was not found among different treatments, especially in AG groups with high dietary energy. That might be due to the combined effect of dietary energy and feed intake (Johnsen et al., 2011). Increased diet energy might improve carotenoid uptake and deposition in salmon muscle while quantity of feed intake could also have the same effect. Restrictive feeding is believed to have the positive effect on the utilization of carotenoids (Ytrest?yl et al., 2004; Johnsen et al., 2011). However, no significant difference was obtained in the present study (26.67–29.33), which may be due to the experimental design based on high feed consumption between two satiation groups during the feeding period.

Fillet firmness has been reported as one of the main flesh characteristics in Atlantic salmon. The effects of several factors contributing to fillet firmness are hydroxylysyl pyridinoline (PYD) cross-links ＞ water content＞ a-i HYP ＞ muscle fibre density (Hagen et al., 2007), and a-i HYP in muscle has been hypothesized to reflect fillet firmness in Atlantic salmon (Johnsen et al., 2011). Present study revealed that feed type with different dietary energy significantly affected firm texture through HYP concentrations (206.06–417.35). On the other hand, restricted feeding had positive effects on a-i HYP concentration among all feed types. This was particularly obvious in SS80% group, with 36% higher than 100% satiation group. This result was in line with another research of feeding regime effects (Johnsen et al., 2011).

Vitamin E is the primary lipid-soluble antioxidant in vertebrates, and the retention rates are lower in fish with less lipids and polyunsaturated fatty acids (PUFAS) in flesh (Bell et al., 1998). In the present study, vitamin E levels in salmon fed AG and HY groups (with higher diet energy) were higher than other two feed types, both expressed as mg kg-1muscle and mg kg-1flesh protein. Besides, satiation degree also had significant effects on vitamin E concentrations, with lower VElevel in lower satiation degree. This result might indicate less vitamin E had been spent as an antioxidant due to reduced feed intake and the elevated vitamin E (α-tocopherol) level in restricted feeding group will reduce lipid oxidation during further processing.

Liquid loss is calculated for liquid holding capacity and is a main flesh quality, which has great consequences for fish further processing and consumers’ acceptance (Elvevoll et al., 1996). Liquid loss is affected by muscle changes in chemical composition and microstructure (L?je et al., 2007), and can be further affected by pH, storage time and temperature (Hagen et al., 2008). In the present study, liquid loss is different among all treatments and higher diet energy groups (AG and HY) seemed to have lower data. It also did not correlate to HYP. The muscle pH also had no effect on liquid loss. Previous study has emphasized that several other factors affect liquid loss in salmon, but the mechanism is unclear. However, 100% group showed higher liquid loss and muscle pH and this result was in line with a previous research (M?rk?re et al., 2008).

Rheological measurements of viscosity, elastic modulus and structural resistance were used to evaluate fecal stability (Brinker, 2007). Previous studies have demonstrated that fish size and developmental stage have important effects on physical characteristics of fecal waste (Muir, 1981; Summerfelt, 1999; Maillard et al., 2005). In RAS, fish size and flow rates were key factors influencing feces characteristics (Franco-Nava et al., 2003). There also existed studies with no such dependency (Chen et al., 1993; Merino et al., 2007). Differences among fecal properties may be the results of the changes in digestive physiology of fish or the composition of the diets or a combination of them (Brinker, 2009). Simultaneously, fish digestive physiology is related to fish size and dietary composition (Guillaume et al., 2001). Thus, small differences in level or quality of raw materials might have major influence on digestibility and led to different fecal properties. Four diets used in the present study were from four manufacturers with different raw materials. HY groups showed better growth and the fish with higher body weight showed worse fecal viscosity. This result was in line with study on rainbow trout (Brinker, 2009).

5 Conclusions

The present study showed that feed type and satiation degree affected growth, mortality and dry matter ADC in Atlantic salmon. Both feed and feeding regime have impacts on protease in stomach and gut and lipase in gut, whilst gut amylase was only be improved by increased feeding level. It is evident that the vitamin E, concentration of mature collagen cross-links (HYP), liquid loss and pH in muscle were affected by feed type and satiation degree, while flesh color was not affected significantly.Although no significant difference was found in the present trial, satiation degree still appeared to increase feces stability, which needs further study.

Acknowledgements

This research was supported by the National Key Technologies R&D Program (2011BAD13B04), the earmarked fund for Modern Agro-industry Technology Research System (CARS-48), and the National Natural Science Foundation of China (No. 31240012). The authors would like to thank Shandong Oriental Ocean Sci-Tech Co., Ltd., for the support during the experiment.

AOAC, 2003. Official Methods of Analysis of the Association of Official Analytical Chemists. 17th edition. AOAC International, Gaithersburg, MD, 684pp.

Azevedo, P. A., Leeson, S., Cho, C. Y., and Bureau, D. P., 2004a. Growth and feed utilization of large size rainbow trout (Oncorhynchus mykiss) and Atlantic salmon (Salmo salar) reared in freshwater: Diet and species effects, and responses over time. Aquaculture Nutrition, 10: 401-411.

Azevedo, P. A., Leeson, S., Cho, C. Y., and Bureau, D. P., 2004b. Growth, nitrogen and energy utilization of juveniles from four salmonid species: Diet, species and size effects. Aquaculture, 234: 393-414.

Bell, J. G., McEvoy, J., Webster, J. L., McGhee, F., Millar, R. M., and Sargent, J. R., 1998. Flesh lipid and carotenoid composition of Scottish farmed Atlantic salmon (Salmo salar). Journal of Agricultural and Food Chemistry, 46: 119-127.

Bendiksen, E. ?., Berg, O. K., Jobling, M., Arnesen, A. M., and M?s?val, K., 2003. Digestibility, growth and nutrient utilisation of Atlantic salmon parr (Salmo salar L.) in relation to temperature, feed fat content and oil source. Aquaculture, 224: 283-299.

Bendiksen, E. A., Johnsen, C. A., Olsen, H. J., and Jobling, M., 2011. Sustainable aquafeeds: Progress towards reduced reliance upon marine ingredients in diets for farmed Atlantic salmon (Salmo salar L.). Aquaculture, 314: 132-139.

Bjerkeng, B., Refstie, S., Fjalestad, K. T., Storebakken, T., R?dbotten, M., and Roem, A. J., 1997. Quality parameters of the flesh of Atlantic salmon (Salmo salar) as affected by dietary fat content and full-fat soybean meal as a partial substitute for fish meal in the diet. Aquaculture, 157: 297-309.

Brinker, A., 2009. Improving the mechanical characteristics of faecal waste in rainow trout: The influence of fish size and treatment with a non-starch polysaccharide (guar gum). Aquaculture Nutrition, 15: 229-240.

Burr, G. S., Wolters, W. R., Schrader, K. K., and Summerfelt, S. T., 2012. Impact of depuration of earthy-musty off-flavors on fillet quality of Atlantic salmon, Salmo salar, cultured in a recirculating aquaculture system. Aquacultural Engineering, 50: 28-36.

Chen, S., Timmons, M. B., Aneshansley, D. J., and Bisogni Jr., J. J., 1993. Suspended solids characteristics from recirculating aquacultural systems and design implications. Aquaculture, 112: 143-155.

Crampton, V. O., Nanton, D. A., Ruohonen, K., Skjervold, P. O., and El-Mowafi, A., 2010. Demonstration of salmon farming as a net producer of fish protein and oil. Aquaculture Nutrition, 16: 437-446.

Ebling, M. E., 1968. The Dumas method for nitrogen in feeds. Journal-Association of Official Analytical Chemists, 51: 766-770.

Einen, O., and Roem, A. J., 1997. Dietary protein/energy ratios for Atlantic salmon in relation to fish size: growth, feed utilization and slaughter quality. Aquaculture Nutrition, 3: 115-126.

Elvevoll, E. O., S?rensen, N. K., ?sterud, B., Ofstad, R., and Martinez, I., 1996. Processing of marine foods. Meat Science, 43: 265-275.

FAO, 2011. FishStat Plus-Universal software for fishery statistical time series [EB/OL]. http: //www.fao.org/fishery/statistics/software/fishstat/en.

Fauré, A., and Labbé, L., 2001. Feeding practices. In: Nutrition and Feeding of Fish and Crustaceans. Guillaume, J., et al., eds., Praxis Publishing Ltd., Chichester, 339-350.

Franco-Nava, M. A., Blancheton, J. P., Deviller, G., Charrier, A., and Le-Gall, J. Y., 2003. Effect of fish size and hydraulic regime on particulate organic matter dynamics in a recirculating aquaculture system: Elemental carbon and nitrogen approach. Aquaculture, 239: 179-198.

Goddardm, J. S., and Mclean, E., 2001. Acid-insoluble ash as an inert reference material for digestibility studies in tilapia, Oreochromis aureus. Aquaculture, 194: 93-98.

Gatlin, D. M., Barrows, F.T., Brown, P. K., Dabrowski, T. G., Gaylord, R. W., Hardy, E., Herman, G., Hu, ?., Krogdahl, R., Nelson, K., Overturf, M., Rust, W., Sealey, D., Skonberg, E. J., Souza, D., Stone, R., and Wilson, E., 2007. Wurtele Expanding the utilization of sustainable plant products in aquafeeds: A review. Aquaculture Research, 38: 551-579

Guillaume, J., Kaushik, S., Bergot, P., and Métailler, R., 2001. Nutrition and Feeding of Fish and Crustaceans. Praxis Publishing Ltd., Chichester, 201-206.

Hagen, ?., Solberg, C., and Johnston, I. A., 2008. Activity of aspargate (cathepsin D), cysteine proteases (cathepsins B, B+L, and H), and matrix metallopeptidase (collagenase) and their influence on protein and water-holding capacity of muscle in commercially farmed Atlantic halibut (Hippoglossus hippoglossus L.). Journal of Agricultural and Food Chemistry, 56: 5953-5959.

Hagen, ?., Solberg, C., Sirnes, E., and Johnston, I. A., 2007. Biochemical and structural factors contributing to seasonal variation in the texture of farmed Atlantic halibut (Hippoglossus hippoglossus L.) flesh. Journal of Agricultural and Food Chemistry, 55: 5803-5808.

Hillestad, M., Johnsen, F., Austreng, E., and ?sg?rd, T., 1998. Long-term effects of dietary fat level and feeding rate on growth, feed utilization and carcass quality of Atlantic salmon. Aquaculture Nutrition, 4: 89-97.

Houlihan, D. F., Hall, S. J., Gray, C., and Noble, B. S., 1988. Growth rate and protein turnover in Atlantic cod, Gadus morhua. Canadian Journal of Fisheries and Aquatic Sciences, 45: 951-964.

Jensen, C., Birk, E., Jokumsen, A., Skibsted, L. H., and Bertelsen, G., 1998. Effect of dietary levels of fat, α-tocopherol and astaxanthin on colour and lipid oxidation during storage of frozen rainbow trout (Oncorhynchus mykiss) and during chill storage of smoked trout. Zeitschrift für Lebensmitteluntersuchung und -Forschung A, 207: 189-196.

Jobling, M., 1994. Fish Bioenergetics. Chapman and Hall, London, 309pp.

Johansen, S. J. S., Ekli, M., and Jobling, M., 2002. Is there lipostatic regulation of feed intake in Atlantic salmon Salmo salar L.? Aquaculture Research, 33: 515-524.

Johansen, S. J. S., Ekli, M., Stangnes, B., and Jobling, M., 2001. Weight gain and lipid deposition in Atlantic salmon, Salmo salar, during compensatory growth: Evidence for lipostatic regulation? Aquaculture Research, 32: 963-974.

Johansen, S. J. S., Sveier, H., and Jobling, M., 2003. Lipostatic regulation of feed intake in Atlantic salmon Salmo salar L. defending adiposity at the expense of growth? Aquaculture Research, 34: 317-331.

Johnsen, C. A., Hagen, ?., Adler, M., J?nsson, E., Kling, P., Bickerdike, R., Solberg, C., Bj?rnsson, B. T., and Bendiksen, E. ?., 2011. Effects of feed, feeding regime and growth rate on flesh quality, connective tissue and plasma hormones in farmed Atlantic salmon (Salmo salar L.). Aquaculture, 318: 343-354.

Johnsen, C. A., Hagen, ?., and Bendiksen, E. ?., 2011. Longterm effects of high-energy, low fish meal feeds on growth and flesh characteristics of Atlantic salmon (Salmo salar L.). Aquaculture, 312: 109-116.

Kunitz, M., 1947. Crystalline soybean trypsin inhibitor. Journal of General Physiology, 30: 291-310.

Le Moullac, G., Klein, B., Sellos, D., and Van Wormhoudt, A., 1997. Adaptation of trypsin, chymotrypsin and a-amylase to casein level and protein source in Penaeus vannamei (Crustacea, Decapoda). Journal of Experimental Marine Biology and Ecology, 208: 107-125.

Li, J. Q., Lin, J. B., Zhu, Q. G., Ma, Y. M., Mei, J. L., and Qiu, M. L., 2005. Effects of different energy-protein ratio in the diet on the activities of digestive enzymes for flounder (Paralichthys olivaceus). Journal of Jimei University, 10: 296-299 (in Chinese).

L?je, H., Jensen, K. N., Hyldig, G., Nielsen, H. H., and Nielsen, J., 2007. Changes in liquid-holding capacity, water distribution and microstructure during chill storage of smoked salmon. Journal of the Science of Food and Agriculture, 87: 2684-2691.

Maillard, V. M., Boardman, G. D., Nyland, J. E., and Kuhn, D. D., 2005. Water quality and sludge characterization at raceway-system trout farms. Aquacultural Engineering, 33: 271-284.

Martins, C. I. M., Eding, E. H., Verdegem, M. C. J., Heinsbroeka, L. T. N., Schneiderc, O., Blanchetond, J. P., Roque d’Orbcasteld, E., and Verretha, J. A. J., 2010. New developments in recirculating aquaculture systems in Europe: A perspective on environmental sustainability. Aquacultural Engineering, 43: 83-93.

McMillan, D. N., and Houlihan, D. F., 1989. Short-term responses of protein synthesis to re-feeding in rainbow trout. Aquaculture, 79: 37-46.

Merino, E. G., Piedrahita, R. H., and Conklin, D. E., 2007. Settling characteristics of solids settled in a recirculating system for California halibut (Paralichthys californicus) culture. Aquacultural Engineering, 37: 79-88.

Miller, M. R., Nichols, P. D., and Carter, C. G., 2008. N-3 Oil sources for use in aquaculture-alternatives to the unsustainable harvest of wild fish. Nutrition Research Reviews, 21: 85-96.

M?rk?re, T., Mazo, T. P. I., Tahirovic, V., and Einen, O., 2008. Impact of starvation and handling stress on rigor development and quality of Atlantic salmon (Salmon salar L). Aquaculture, 277: 231-238.

Muir, J. F., 1981. Economic aspects of waste treatment in fish culture. In: Report of the Eifac Workshop on Fish-Farm Effluents. Alabaster, J. S., ed., FAO, Silkeborg, 123-135.

Naylor, R. L., Hardy, R.W., Bureau, D. P., Chiu, A., Elliott, M., Farrell, A. P., Forster, I., Gatlin, D. M., Goldburg, R. J., Hua, K., and Nichols, P. D., 2009. Feeding aquaculture in an era of finite resources. Proceedings of the National Academy of Science, 106: 15103-15110.

Norwegian Standard, 1994. Atlantisk laks. Mailing av forge og fett. NS 9402, Norsk Standard, Oslo, Norway, 5pp.

Ofstad, R., Kidman, S., Myklebust, R., and Hermansson, A. M., 1993. Liquid-holding capacity and structural changes during heating of fish muscle: Cod (Gadus morhua L.) and salmon (Salmo salar). Food Structure, 12: 163-174.

Pratoomyot, J., Bendiksen, E. A., Bell, J. G., and Tocher, D. R., 2010. Effects of increasing replacement of dietary fishmeal with plant protein sources on growth performance and body lipid composition of Atlantic salmon (Salmo salar L.). Aquaculture, 305: 124-132.

Rick, W., and Stegbauer, H. P., 1984. Alfa-amylase: Measurement of reducing groups. In: Methods of Enzymatic Analysis, Enzymes. 3rd edition. Bergermeyer, H. U., and Grab, M., eds., ChemieVerlag, Weinhein, 5: 885-889.

Shao, Q. J., Su, X. F., and Xu, Z. Z., 2004. Effects of dietary protein levels on growth performance and digestive enzyme activities of jade perch Scortum bacoo. Journal of Zhejiang University, 30: 553-556 (in Chinese).

Solberg, C., 2004. Influence of dietary oil content on the growth and chemical composition of Atlantic salmon (Salmo salar). Aquaculture Nutrition, 10: 31-37.

Summerfelt, S. T., 1998. An integrated approach to aquaculture waste management in flowing water systems. Proceedings of 2nd International Conference on Recirculating Aquaculture, Roanoke, 87-97.

Summerfelt, S. T., 1999. Waste-handling systems. In: CIGR Handbook of Agricultural Engineering. Wheaton, F. W., ed., American Society of Agricultural Engineers, Maryland, 309-350.

Tacon, A. G. J., and Metian, M., 2008. Global overview on the use of fish meal and fish oil in industrially compounded aquafeeds: Trends and future prospects. Aquaculture, 285: 146-158.

Torstensen, B. E., Espe, M., Sanden, M., Stubhaug, I., Waagb?, R., Hemre, G. I., Fontanillas, R., Nordgarden, U., Hevr?y, E. M., Olsvik, P., and Berntssen, M. H. G., 2008. Novel production of Atlantic salmon (Salmo salar) protein based on combined replacement of fish meal and fish oil with plant meal and vegetable oil blends. Aquaculture, 285: 193-200.

Tucker Jr., J. W., 1998. Marine Fish Culture. Kluwer Academic Publishers, Boston, 750pp.

Turchini, G. M., Torstensen, B. E., and Ng, W. K., 2009. Fish oil replacement in finfish nutrition. Reviews in Aquaculture, 1: 10-57.

Unger, J., and Brinker, A., 2013. Feed and treat: What to expect from commercial diets. Aquacultural Engineering, 53: 19-29.

Windell, J. T., Foltz, J. W., and Sarokon, J. A., 1978. Methods of fecal collection and nutrients leaching in digestibility studies. The Progressive Fish-Culturist, 40: 51-55.

Ytrest?yl, T., Coral-Hinostroza, G., Hatlen, B., Robb, D. H. F., and Bjerkeng, B., 2004. Carotenoid and lipid content in muscle of Atlantic salmon, Salmo salar, transferred to seawater as 0+ or 1+ smolts. Comparative Biochemistry and Physiology Part B: Biochemistry and Molecular Biology, 138: 29-40.

(Edited by Qiu Yantao)

(Received December 21, 2013; revised February 19, 2014; accepted May 25, 2015)

J. Ocean Univ. China (Oceanic and Coastal Sea Research)

DOI 10.1007/s11802-015-2565-5

ISSN 1672-5182, 2015 14 (5): 849-857

http://www.ouc.edu.cn/xbywb/

E-mail:xbywb@ouc.edu.cn

* Corresponding author. yinliu@qdio.ac.cn