Dietary arachidonic acid affects immune function and fatty acid composition in cultured rabbitfish Siganus rivulatus

Fish & Shellfish Immunology 68 (2017) 46e53

Contents lists available at ScienceDirect

Fish & Shellfish Immunology journal homepage: www.elsevier.com/locate/fsi

Full length article

Dietary arachidonic acid affects immune function and fatty acid composition in cultured rabbitfish Siganus rivulatus Sagar Nayak a, William Koven b, Iris Meiri b, Inna Khozin-Goldberg a, Noah Isakov c, Mohammad Zibdeh d, Dina Zilberg a, * a

The Jacob Blaustein Institutes for Desert Research, Ben-Gurion University of the Negev, Midreshet Ben-Gurion, Israel Israel Oceanographic and Limnological Research, National Centre for Mariculture, Eilat 8812, Israel The Shraga Segal Department of Microbiology, Immunology and Genetics, Ben-Gurion University of the Negev, Beer-Sheva, Israel d Marine Science Station, University of Jordan/Yarmouk University, Aqaba, Jordan b c

a r t i c l e i n f o

a b s t r a c t

Article history: Received 12 April 2017 Received in revised form 25 June 2017 Accepted 1 July 2017 Available online 4 July 2017

The marbled spinefoot rabbitfish (Siganus rivulatus) is an economically valuable fish species that has potential for commercial production in aquaculture. To overcome challenges in its sustainable production, a formulated diet is required for imparting health and robustness. This study evaluates the effect of dietary supplementation with arachidonic acid (ARA; 20:4n-6) on growth, survival, immune function and fatty acid composition of red blood cells (RBCs) in rabbitfish. We conducted two feeding trials using juvenile fish (to evaluate growth and survival) and adults (to evaluate immune function and fatty acid incorporation). Fish were fed diets supplemented with three different levels of ARA (in % of total fatty acids): 0.6 (unsupplemented control), 2.6 (moderate) and 4.7 (high). The fish fed with moderate ARA levels exhibited improved (p < 0.05) growth over the control and the high ARA level groups. During an outbreak of Streptococcus iniae, fish fed with moderate ARA survived significantly (p < 0.05) better (89%) than the control and the high ARA groups (59% and 48%, respectively). Moderate ARA supplementation resulted in elevated lysozyme and complement levels in the plasma of rabbitfish. A significant increase in the total serum immunoglobulin levels was observed in both the medium and the high ARA level groups; however, a decrease in antiprotease activity was recorded in the supplemented groups as compared to the control. Fatty acid analysis in fish red blood cells revealed a significant (p < 0.05) increase in the proportion of ARA of total fatty acids in the groups fed with the medium and the high ARA level diets (9.5% and 11.2%, respectively, compared to 7.1% in the control). Concomitantly, there was a decrease in the proportion of eicosapentaenoic acid (EPA; 20:5n-3), dihomo-g linolenic acid (DGLA; 20:3n-6) and several 18-carbon unsaturated fatty acids in these groups. In conclusion, ARA in rabbitfish feeds improved growth, survival as well as innate and acquired humoral immune functions. Thus ARA supplementation in the diet of this species could be a valuable step towards establishing the commercial culture of rabbitfish. © 2017 Elsevier Ltd. All rights reserved.

Keywords: Arachidonic acid Rabbitfish Siganus rivulatus Innate immunity Fatty acid composition

1. Introduction Rabbitfish (Siganids) or spinefoot are a family of herbivorous marine fishes that are native to the Red Sea. The markets for Siganus spp. are well founded in Jordan, Egypt and other countries in the Middle East and around the Mediterranean basin. The marbled spinefoot rabbitfish (Siganus rivulatus) is greatly valued in the Saudi

* Corresponding author. E-mail address: [email protected] (D. Zilberg). http://dx.doi.org/10.1016/j.fsi.2017.07.003 1050-4648/© 2017 Elsevier Ltd. All rights reserved.

Arabian market and was selected as an important species in its national mariculture development program [1]. Siganus spp represent one third of all imported species supplying the national consumption of Jordanian aquatic products [2]. Apart from its promising market demand, the marbled spinefoot rabbitfish (Siganus rivulatus) is an excellent candidate for regional production as the culture of this herbivorous species would be sustainable, due to the fact that no fish meal or oil is required in the diet [1,2]. In research carried out at the National Center for Mariculture in Eilat, Israel, major advances related to this species' reproduction in

S. Nayak et al. / Fish & Shellfish Immunology 68 (2017) 46e53

captivity and grow out to the juvenile stages have been achieved (MERC Final Report, 2013). However, reliable and predictable production of rabbitfish remains elusive. Moreover, an effective growout dietary formulation would need to not only support optimal growth, but also promote disease resistance and robustness in juvenile fish. Within the animal kingdom, dietary lipids are considered a key source of energy, which also contribute essential fatty acids (EFA) and phospholipids for cell membrane synthesis and gene expression. The role of the EFA fatty acid arachidonic acid (ARA; 20:4n-6) in promoting growth, survival and stress resistance has been well documented [3e6]. ARA has a pronounced effect on immune functions through the generation of inflammatory mediators of the eicosanoid family, including prostaglandins, leukotrienes, and related metabolites that regulate the activity of inflammatory cells and the production of cytokines and various defense factors [7e9]. Thus, dietary lipids may have a significant impact on the immune system and the health status of fish. Published data on the physiology and immune system of rabbitfish are scarce. A newly described anti-parasitic protein from Siganus oramin was shown to promote immune resistance against the protozoan parasite, Cryptocaryon irritans [10,11]. To the best of our knowledge, no studies have been carried out on the potential modulatory effects of ARA on the immune system of rabbitfish. Furthermore, only very few studies have examined the fatty acid composition in fish red blood cells following a dietary supplementation with an n-6 long chain polyunsaturated fatty acid (LCPUFA). The advantage of performing the analysis on red blood cells is that the incorporation of dietary PUFA, as a function of time, can be evaluated in living fish without the need to sacrifice them. The aims of the present study were to examine the effect of dietary ARA on rabbitfish (1) growth and survival and (2) nonspecific and adaptive immune function through the evaluation of lysozyme, complement ACH50 and total immunoglobulin, and (3) by using a multivariate approach, to elucidate the interactions/relationships between the immune functions and the red blood cells' fatty acid profile. 2. Material and methods 2.1. Arachidonic acid feeding experiments Two separate feeding trials of dietary supplementation with ArA were conducted to study the effect of this essential fatty acid on the growth, survival and innate immune function of juvenile and adult rabbitfish. Arasco™ oil (DSM, Kaiseragst, Switzerland), which contains a minimum of 38% ARA of total fatty acids (TFA), was used as a source of this n-6 LCPUFA in the food preparation. The experimental food for the trials with three different levels of ARA (0.6, 2.6, and 4.7% of TFA) and a pellet size of 1.6 mm was produced at the National Center for Mariculture (NCM) in Eilat, Israel. 2.1.1. Trial 1: effect of diets containing different concentrations of ARA on rabbitfish growth and survival Approximately 400 juvenile rabbitfish were caught by seine net in November 2010 in a seawater lagoon close to the NCM in Eilat, Israel. The fish were maintained for about three weeks in a 700-L square tank, fed for about a week with fresh Ulva lactuca, and then weaned onto the control pelleted diet having the lowest level of ARA (0.6% of TFA). After weaning, 360 similar-sized fish (10e25 g) were distributed among twelve 200-L conical tanks (30 fish per tank) in a flow-through seawater system (40‰, 25
C). This allowed a 65 days feeding trial, testing three ARA pelleted diets (0.6, 2.6, and 4.7% ARA of TFA) in replicates of four tanks per treatment. After weighing the fish in each tank (homogeneity of

47

variance, ANOVA > 0.05), the fish were fed daily with their respective diets at 1e3%. Diet ration level was adjusted according to the water temperature, which was initially 25
C (at the beginning of the trial on Dec 8th), gradually dropping during Jan, reaching 22
C in Feb, during the last 10 days of the trial. Feeding was initially 3% of fish weight, decreasing with the decrease in temp, reaching 1% when the water temp reached 22
C. The fish were weighed approximately every two weeks, where all fish in each of the tanks were weighed together, and feeding rations were adjusted according to the percent biomass. Fish were monitored for behavior, morbidity and mortality throughout the experiment. At appearance of sick or moribund fish, wet mounts of skin, gills and internal organs were microscopically examined. In addition, internal organs of these fish were aseptically streaked on plates containing brain heart infusion; BHI and tripton sugar agar (TSA). Isolated bacteria were gram stained and identified, based on biochemical characteristics, using the API test kit (API 20 Strep, Biomeriuex, USA). 2.1.2. Trial 2: effects on immune function and lipid composition Forty-eight three-year-old F1 rabbitfish (225 ± 50 g), which were progeny of the wild-caught fish that were used in Trial 1, were reared at the NCM and stocked in three separate square 700-L tanks (16 fish per tank) supplied with a continuous flow of seawater from the Red Sea (40‰, 25
C). Fish were acclimated for three weeks, during which time they were fed the non-supplemented control (0.6% ARA of TFA) diet. Experimental diets were identical in their protein, lipid and energy contents but differed in their levels of ARA content (0.6, 2.6 and 4.7% of TFA) and were referred to as control, medium and high ARA, respectively. Fish were fed 2e3 times daily to satiation. One ml of blood was drawn from each fish (using a heparinized syringe) at time 0, 1, 2 and 7 months. Samples were centrifuged at 2000  g for 15 min at 4
C. Plasma was collected into separate 1-ml tubes, and both the plasma and the cell pellet were kept at 80
C for further analysis. 2.2. Analysis of innate immune functions in plasma 2.2.1. Lysozyme assay A lysoplate assay, based on lysis of the Micrococcus lysodeikticus bacterium on agar plates, was carried out as described by Lie et al. [12] with some modifications. The diameters of the clear lysed plaques were measured, and lytic activity was calculated by comparison with the results of a standard curve obtained using predetermined concentrations of hen egg white lysozyme. Plates were prepared by adding M. lysodeikticus at 150 mg/ml to a 1% suspension of agarose in phosphate citrate buffer (0.1 M, pH 5.8), and mixing the suspension to obtain homogeneity. The suspension was poured into petri dishes (10  10 cm) and allowed to solidify at room temperature. Wells were punched into the solidified plates using a plunger of 3-mm diameter. Standards were prepared using hen egg white lysozyme (Sigma, L-6876, EC 3.2.1.17) at concentrations of 12e1000 mg/ml in PBS (0.05 M, pH 6.2). Eight ml of the tested samples or standards were added to triplicate wells. After 24 h of incubation at 22
C, the zone of clearance around the wells was measured. Results were expressed as the diameter of the clear zone proportional to the log of the lysozyme concentrations (12e1000 mg/ml). 2.2.2. Antiprotease assay Antiprotease activity in plasma samples was measured according to Magnadottir et al. [13]. Briefly, 20 ml of plasma was incubated with 20 ml of trypsin (5 mg/ml) in a 1.5-mL Eppendorf tube for 5 min. Maximal proteolytic activity (100%) was obtained by mixing 200 ml PBS with 250 ml azocasein (protein substrate; at 20 mg/ml).

48

S. Nayak et al. / Fish & Shellfish Immunology 68 (2017) 46e53

The negative control (blank) reaction mixture included PBS instead of serum and trypsin. Samples were incubated at 22
C for 1 h, and reactions were terminated by the addition of 500 ml of 10% tricarboxylic acid for 30 min, followed by centrifugation at 6000  g for 5 min and separation of the supernatant. Finally, 100 ml of the supernatants were added to the wells of a 96-well plate along with 100 ml of 1N NaOH, and absorbance at 405 nm was measured using a Tecan Sunrise™ microplate reader (Salzburg, Austria). Percentage inhibition of trypsin activity was calculated using the formula: % inhibition ¼ (A1-A2/A1)  100 where A1 ¼ OD of maximal trypsin activity (without serum); A2 ¼ OD of trypsin activity with the serum sample-serum blank. 2.2.3. Alternative complement (ACH50) assay ACH50 was determined according to Sunyer and Tort (1995), with a few modifications. Briefly, rabbit red blood cells (RRBCs) were used as target cells in the presence of gelatin veronal buffer (EGTA-GVB). Individual 100-ml aliquots (twofold dilutions) of a serially diluted serum in EGTA-Mg2þ GVB (at a range of 1:16e1:1024) were mixed with 30 ml of RRBC suspension (4  108 cells ml1). Maximal (100%) lysis was obtained by mixing 30 ml of RRBC with 1100 ml of double distilled water (DDW), and blank samples were obtained by mixing the RRBCs with EGTAMg2þ GVB. The reaction mixtures were incubated at 23
C with gentle shaking for 2 h. The hemolytic reaction was stopped by adding 1000 ml of buffer containing 10 mM EDTA. The mixtures were centrifuged at 1600  g for 10 min at 4
C, and the optical density of the supernatants was determined at 414 nm using a microplate reader. The extent of hemolysis was calculated according to Sunyer and Tort (1995) and presented as the number of ACH50 units ml1 that provided 50% lysis. 2.3. Development of capture ELISA and total immunoglobulin analysis Antibodies directed against Siganus immunoglobulins were prepared in mice and in rabbits according to Sharon et al. [14]. A capture ELISA assay was designed and calibrated for estimation of the total immunoglobulin in fish plasma. Briefly, flat-bottom ELISA microtiter plate wells (96-well, Nunc-Maxisorp, Roskilde, Denmark) were each coated with 100 ml (diluted 1:10 in Trisbuffered saline, pH 7.6) of rabbit anti-siganus IgM antibodies by overnight incubation at 4
C. Wells were washed three times with TBS-T (TBS þ 0.05% Tween-20) and once with TBS. This washing protocol was applied after each addition of antigen or antibody throughout the assay, unless differently indicated. Wells were then blocked with 200 ml of 5% skimmed milk in TBS, pH 7.6 (blocking solution) for 2 h at RT, followed by one more wash in TBS. Tested rabbitfish sera were diluted 1:10 in TBS-T þ 3% skimmed milk (dilution buffer), added to rabbit anti-Siganus IgM antibodyprecoated wells, in triplicates, and incubated for 2 h at RT, followed by washing. One hundred ml of mouse anti-siganus antiserum, diluted 1:3000 in dilution buffer, was added to each well and incubated for an additional 2 h at RT. The wells were washed, and 100 ml of HRP-conjugate goat anti-mouse IgG (Bio-Rad laboratories Inc., Hercules, CA, USA), diluted 1:3000 in TBS-T þ 3% skimmed milk, was added to each well and incubated for another 2 h at room temperature. Wells were washed followed by addition of 100 ml of TMB peroxidase substrate (3, 30 , 5, 5’-tetramethylbenzidine solution plus hydroxide; Bio-Rad) 30 min of incubation at RT in the dark. Finally, the reaction was terminated by adding 50 ml of 1 N H2SO4, and absorbance was recorded at 450 nm (Tecan Sunrise™, Salzburg, Austria).

2.4. Analysis of fatty acid composition in red blood cells Rabbitfish red blood cells were collected two months postinitiation of the feeding experiments, stored at 80
C and then analyzed for total fatty acid profile. Red blood cells were freezedried overnight, and 10 mg of the resulting powder was transmethylated by the addition of a 2-ml solution containing 2% H2SO4 in dry methanol (v/v) and 0.01% BHT (butylated hydroxytoulene), under an argon atmosphere. The fatty acid C17:0 was used as an internal standard. The reaction mixture was heated at 80
C for 90 min in a dry sand bath, and terminated by the addition of 1 ml H2O. Fatty acid methyl esters (FAMEs) were extracted with 1 ml n-hexane, and the mixture was then vortexed and centrifuged at 1000  g for 5 min at RT. The supernatant was collected and evaporated under nitrogen flow to dryness. GC analysis of FAMEs was carried out on a Trace Ultra gas chromatograph (Thermo) equipped with a flame ionization detector, a programmable temperature detector and a capillary column (ZB WAXplus, Phenomenex, USA). FAMEs were identified by co-chromatography with authentic standards (Sigma-Aldrich, Israel) and a fish oil commercial standard. The relative proportions of individual fatty acids were calculated by dividing the individual GC peak areas by total peak area of all detected fatty acids excluding internal standards. The limit of quantification was 0.2% of total fatty acids and detection was ~0.5 mg. 2.5. Statistics Linear regression, comparison of slopes and goodness of curve fit (r2) were performed (Prism 5, Graph Pad Software Inc.) on growth and survival values from trial 1. Results of immune analyses were compared by one way and two way analyses of variance (ANOVA), using the SigmaPlot 13 Software (Systat Software Inc.,). Outliers were identified using residual analysis using Origin Pro (Origin Lab Inc.). If ANOVA of values were significant (p < 0.05), Tukey or Dunn's post hoc tests for multiple pairwise comparisons was then carried out. Survival was analyzed by the Kaplan Meyer Survival Analysis (Sigma Plot 13). A multivariate analysis of data by principal component analysis (PCA) was carried out using PAST 3.15 (free program available at https://folk.uio.no/ohammer/past/), to analyze the extent to which fatty acid and immune profiles in fish differed between dietary treatments of ARA. The analysis was carried out using individual fatty acid proportions and values obtained from immune analyses as input variables. The fatty acid composition data (percentage values) were transformed (arcsin) and the variables were normalized (as the variables had different scales and units). Variables were then used to generate a PCA correlation matrix and biplots. 3. Results 3.1. Effects of diets containing different ARA concentrations on growth and survival of juvenile rabbitfish Fish fed with the medium ARA-supplemented diet (2.6% of TFA) exhibited a linear (r2 ¼ 0.92) wet weight gain (g) per day (ca. 2% body weight per day) which more than doubled their initial weight during the study and was significantly (p ¼ 0.027) better than the growth rates exhibited in the control (0.6% ARA of TFA) and the high ARA-supplemented (4.6% of TFA) fish (Fig. 1). Linear regression analysis revealed r2 values of 0.8, 0.98 and 0.2 for the control, medium and high supplemented groups, demonstrating the most linear growth in the medium-supplemented diet. The growth curve slopes (0.2, 0.3 and 0.1 for the control, medium and high supplemented groups) were significantly different (p < 0.027), suggesting

S. Nayak et al. / Fish & Shellfish Immunology 68 (2017) 46e53

49

Fig. 1. Effects of different ARA diets on average weight gain of juvenile rabbitfish. Values within brackets represent the regression coefficient (r2) while the rate or slope of weight gain was significantly (p < 0.05) higher in the moderate ARA diet than in the low and high ARA diets (n ¼ 4 tanks, 30 fish per tank).

significantly higher growth rate for the medium, 2.6% ARA of total TFA-supplemented group. About two weeks into the study, there was an outbreak of Steptococcus iniae, which was identified in a number of fish from the 0.6% and 4.6% ARA treatments, resulting in considerable daily mortality that decimated a number of replicates from these treatment groups. Survival in fish fed the low and high ARA diets decreased significantly (P < 0.05), regressed in linear fashion, while there was not a single confirmed mortality from this disease in fish fed with the 2.6% ARA diet (Fig. 2). Final survival averaged 59% in the low, 87% in the medium, and 49% in the high ARA treatment groups (Fig. 2).

3.2. Effects of different ARA diets on lysozyme levels in rabbitfish plasma Fish fed the moderate ARA-supplemented diet demonstrated a significant (p < 0.05) increase in the plasma lysozyme levels at one month and at seven months, compared to both the control and high ARA groups (Fig. 3). A decrease in lysozyme levels in this group was observed at two months, which was lower than the 1 and 7 month

Fig. 2. Effects of different ARA diets on survival of juvenile rabbitfish. Data represents mean ± SD of percent survival from four replicate tanks (n ¼ 30 per tank). Values within brackets represent the regression coefficient (r2) while the rate or slope of decrease of the average number of fish per tank was significantly (P < 0.05) less in the 2.6% ARA treatment; different lower case letters (a, b, c) denote significant differences in survival over the experimental period (p < 0.05), based on the Kaplan Meyer survival analysis.

Fig. 3. Lysozyme levels in rabbitfish fed the different ARA diets. Values represent average concentrations of lysozyme, in mg/ml ± SEM. The different symbols Ɨ,¥, indicate significant differences (p < 0.05) in lysozyme levels between treatments at a specific time point; the different lower case letters (a, b) indicate significant (p < 0.05) differences within treatments at different time points.

time points. Lysozyme levels in fish fed the high ARA diet significantly (p < 0.05) increased after seven months, compared to levels in the previous months. 3.3. Effects of different ARA diets on antiprotease levels in rabbitfish plasma Both the medium and the high ARA groups of rabbitfish experienced a decrease in plasma antiprotease levels at all sampling times compared to time 0 (Fig. 4). Levels in the medium ARA group were also significantly (p < 0.05) lower than those in the control group at the 2 month time point. The levels of antiprotease in the control group did not significantly change over time (Fig. 4). 3.4. Effects of different ARA diets on ACH50 activity in rabbitfish plasma The complement system comprises a large number of different plasma proteins that play key roles in innate and adaptive

Fig. 4. Determination of antiprotease levels in rabbitfish fed diets with different ARA concentrations, using the trypsin inhibition assay. Values represent percentage of trypsin inhibition measured ± SEM. The different symbols Ɨ,¥ indicate significant differences (p < 0.05) in antitrypsin levels between treatments at a certain specific time point. The different lower case letters indicate significant differences (p < 0.05) within treatments at different time points.

50

S. Nayak et al. / Fish & Shellfish Immunology 68 (2017) 46e53

immunity. The alternative complement activity (ACH50) assay, performed here, is antibody-independent. The results demonstrated that rabbitfish fed with ARA-supplemented diets had increased complement activity at the 2 and 7 month time points, when compared to time 0 or to the 1 month time point, within the same groups (Fig. 5). Complement activity levels in the medium ARA group were higher than those in the control group at the 2 month time point. 3.5. Effects of different ARA diets on total immunoglobulin levels in rabbitfish plasma A capture ELISA assay was developed as part of this study and used to measure plasma immunoglobulin levels in rabbitfish plasma. A significant (p < 0.05) increase in the total plasma immunoglobulin levels was observed in groups fed the medium and high ARA diets at the 2 month time point (Fig. 6). There were no significant changes in the total immunoglobulin levels over time in the control fish.

3.6. Effects of different ARA diets on fatty acid composition in rabbitfish red blood cells Fatty acid analysis in fish RBCs was carried out after two months of feeding with the different experimental diets. The major fatty acids observed in red blood cells were palmitic acid (PA; 16:0), oleic acid (OL; 18:1n-9) and linoleic acid (LNA; 18: 2n-6). Among LCPUFA, ARA and docosahexanoic acid (DHA; 22:6n-3) dominated, while eicosapentanoic acid (EPA; 20:5n-3) occurred at lower levels. The results indicated a significant increase in the proportion of ARA in TFA, reaching 9.5% and 11.2% in the medium and high level groups, respectively, as compared to 7.1% in the control (Table 1). Concomitantly, a decrease in the proportion of EPA, DGLA and several 18-carbon unsaturated fatty acids was observed in the ARAsupplemented groups (Table 1). An increase in 22:4n-6 was measured in the ARA-supplemented groups, although significant values were obtained only in the high ARA groups. Ratios of EPA/ ARA and EPA þ DHA/ARA decreased in the medium and high ARA groups relative to the control. 3.7. Effects of different dietary levels of ARA on immune factors and fatty acid content in rabbitfish red blood cells

Fig. 5. Plasma hemolytic complement activity (ACH50) levels in rabbitfish fed diets with different ARA concentrations. Values represent average units/ml ± STD. The different symbols Ɨ, ¥ indicate significant (p < 0.05) differences in ACH50 levels between treatments at a specific time point. The different lower case letters indicate significant (p < 0.05) differences within treatments at different time points.

PCA loadings and plots were generated based on selected fatty acid variables obtained by analyzing the fatty acid composition in red blood cells (Table 2). Only eigenvalues >1.0 were considered significant elements for data variance (as per Kaiser's rule). The calculated eigenvalues for the first three components were 6.51, 3.10 and 2.19, respectively. The eigenvalues indicated a significant contribution of fatty acid variables to the total variability explained by the PCA. Together PC1 and PC2 explained 53% (36.21% and 17.24%, respectively) of the variance, while PC3 explained only 12.20%. Thus, these three components were sufficient for interpreting the relationship between fatty acid profile, immune functions and dietary ARA supplementation in rabbitfish. The PCA plot revealed that fatty acid compositions were differently affected based on the dietary treatments (Fig. 7). The samples with similar relative fatty acid compositions and immune profiles were located in the same area in the score plot. Variables such as ARA and n-6 fatty acid products (22:4n-6 and 22:5 n-6) were closely clustered indicating positive correlations. Scores from PC1 indicated high contributions from ARA (0.78), 22:4n-6 (0.80), 22:5n-6 (0.08), and DHA (0.70), and moderate loadings from lysozyme (0.27), complement (0.29) and total immunoglobulins (0.37). PC1 also showed highly negative loadings from 18:1n-9 (0.84), 18:2n-6 (0.92), 16:1n-7 (-0.71) and 18:3n-6 (0.91). The loading from these variables indicates that these are responsible for the maximum variance in the data. Similarly, positive contributions from analyzed innate and acquired immune functions were obtained from this analysis. PC2 showed positive loadings from 14:1 (0.70), 16:4 (0.07), and 20:0 (0.76), and negative loadings from DGLA (0.39), EPA (0.64) and DHA (0.45). However, PC3 displayed positive loadings from lysozyme (0.63), complement (0.69), and total Ig (0.61); most of the other variables seemed to have less effect on the variance. 4. Discussion

Fig. 6. The effects of diets with different ARA concentrations on total immunoglobulin levels in rabbitfish plasma. Values represent average OD at 450 nm ± SEM. The different symbols Ɨ, ¥ indicate significant differences (p < 0.05) in immunoglobulin levels between treatments at a specific time point. Different lower case letters indicate significant differences (p < 0.05) within treatments at different time points.

Dietary LC-PUFAs are a major factor in animal diets as these essential fatty acids contribute to tissue membranes, eicosanoid production and gene expression [15]. Research on ARA, which has been shown to directly affect immune function, has been carried out mainly on marine species [16e18]. However, there are very few studies on the ARA requirement in herbivorous species, such as the rabbitfish [19,20]. In the present study, diets with two different

S. Nayak et al. / Fish & Shellfish Immunology 68 (2017) 46e53 Table 1 Fatty acid composition and HUFA (highly unsaturated fatty acids) content in red blood cells of rabbitfish fed diets supplemented with, 2.6 and 4.7% ARA of TFA and the unsupplemented control (0.6% ARA of TFA). Values are presented as mean ± SE, n ¼ 16. Fatty acid

Dietary ARA (% of TFA) 0.6 (control)

2.6

4.7

Fatty acid composition (% of TFA) 14:0 14:1 16:0 16:1n-5 16:1n-7 16:1* 16:2* 16:2* 16:3 16:4 18:0 18:1n-9 18:1n-7 18:2* 18: 2n-6 18:3n-6 18:3n-3 18: 4n-3 20:0 20:1 20:2 20:3n-6 (DGLA) 20:4n-6 (ARA) 20:3n-3 20:4n-3 20:5n-3 (EPA) 22:0 22:1 22:4n-6 22:5n-6 22:5n-3 22:6n-3 (DHA) EPA/AA EPA þ DHA/ARA TFA (mg/mg1) ARA (mg mg1) EPA (mg mg1) DHA (mg mg1)

tr 0.1 ± 0.0 24.7 ± 0.4 0.6 ± 0.1 2.1 ± 0.2 0.2 ± 0.0 0.2 ± 0.0 0.2 ± 0.0 0.3 ± 0.0 0.2 ± 0.0 6.9 ± 0.3 12.7 ± 1.1a 2.0 ± 0.1 0.6 ± 0.0 11.4 ± 0.4a 2.0 ± 0.2a 0.6 ± 0.0 0.2 ± 0.0 0.2 ± 0.0 1.1 ± 0.1 0.5 ± 0.0 2.2 ± 0.1a 7.1 ± 0.5a tr 0.3 ± 0.0 1.2 ± 0.1a tr 0.5 ± 0.1 0.6 ± 0.0a 1.2 ± 0.1 1.8 ± 0.1 17.2 ± 1.4 0.2 2.6 32.1 ± 3.9a 2.0 ± 0.1b 0.4 ± 0.0a 5.0 ± 0.4

0.8 ± 0.1 0.2 ± 0.0 25.0 ± 0.4 0.6 ± 0.0 1.6 ± 0.2 0.2 ± 0.0 0.2 ± 0.0 0.3 ± 0.0 0.3 ± 0.0 0.2 ± 0.0 7.7 ± 0.3 9.5 ± 0.7 ab 1.8 ± 0.1 0.4 ± 0.0 9.2 ± 0.2b 1.3 ± 0.1b 0.5 ± 0.0 0.2 ± 0.0 0.2 ± 0.0 0.9 ± 0.0 0.5 ± 0.0 1.9 ± 0.1ab 9.5 ± 0.4b tr 0.3 ± 0.0 0.9 ± 0.1b tr 0.3 ± 0.0 1.0 ± 0.0a 1.6 ± 0.0 2.0 ± 0.1 21.1 ± 1.1 0.1 2.3 24.0 ± 2.5b 2.2 ± 0.2a 0.2 ± 0.0b 4.8 ± 0.4

0.9 ± 0.1 0.4 ± 0.1 24.9 ± 0.5 0.6 ± 0.0 1.8 ± 0.2 0.2 ± 0.0 0.3 ± 0.0 0.4 ± 0.1 0.4 ± 0.0 0.3 ± 0.0 7.8 ± 0.3 9.9 ± 0.8a 2.0 ± 0.1 0.4 ± 0.0 8.1 ± 0.4b 1.1 ± 0.1b 0.5 ± 0.0 0.2 ± 0.0 0.2 ± 0.0 1.0 ± 0.1 0.5 ± 0.0 1.7 ± 0.1b 11.2 ± 0.8b tr 0.2 ± 0.0 0.8 ± 0.0b tr 0.3 ± 0.0 1.0 ± 0.1b 1.9 ± 0.1 1.9 ± 0.1 19.6 ± 1.0 0.1 1.8 22.6 ± 2.2b 2.4 ± 0.1a 0.2 ± 0.0b 4.3 ± 0.4

51

Table 2 Eigen analysis of the correlation matrix loadings for significant principal components. Variables

PC1

PC2

PC3

14:1 16:1n-7 18:1n-9 18:2 18:2n-6 18:3n-6 20:0 DGLA ARA EPA 22:4n-6 22:5n-6 DHA Lysozyme Complement (ACH50) Antiprotease Total_Ig Eigenvalue % variance

0.379 ¡0.713 ¡0.843 ¡0.772 ¡0.916 ¡0.913 0.136 0.229 0.776 0.152 0.800 0.771 0.701 0.268 0.286 0.010 0.365 6.51 36.21

0.701 0.459 0.378 0.393 0.053 0.112 0.763 0.392 0.084 ¡0.636 0.219 0.292 0.448 0.030 0.199 0.184 0.107 3.10 17.24

0.180 0.141 0.105 0.028 0.064 0.076 0.075 0.432 0.492 0.190 0.307 0.369 0.218 0.634 0.694 0.309 0.612 2.19 12.20

*A total of 14 fatty acid variables and four immune indices were selected for the analyses. PC1, PC2 and PC3 refer to principal components 1, 2 and 3, respectively. The values in the table represent loadings (indicating degree and direction of the relationship of the variables within a principal component). Significant loading values (>0.5) are indicated in bold numbers. (þand -) signs indicate positive and negative correlations, respectively. ARA, arachidonic acid; TFA, total fatty acid; DGLA, dihomo-g-linolenic acid; EPA, Eicosapentaenoic acid; DHA, Docosahexaenoic acid.

*Unidentified isomer of 18:2 and 16:2. a, b Different letters denote differences between treatments at a significance of p < 0.05. ARA, arachidonic acid; TFA, total fatty acid; DGLA, dihomo-g-linolenic acid; EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid; tr, traces.

levels of ARA supplementation and the non-ARA-supplemented control diet were fed to rabbitfish to evaluate the effect of this n6 LCPUFA on growth, survival, the immunological response and RBC fatty acid composition. The innate immune system operates as a first line of defense [13], especially in lower order animals, such as fish. In the present study, we tested the effects of ARA-supplemented diets on innate immune system parameters, such as the plasma lysozyme levels, complement levels, and antiprotease activity in rabbitfish. The results obtained support the assumption that addition of ARA to the rabbitfish diet significantly improves their innate immune functions. It is assumed that the primary reason for the observed effects are due to the fact that ARA is the primary precursor of eicosanoids, which are directly responsible for the regulation and activation of biochemical cascades within the immune system [21e23]. Plasma lysozyme is commonly recognized as a non-specific immune index owing to its mucolytic properties that breakdown the peptidoglycan layer of the bacterial cell wall, causing lysis [24] and activating phagocytosis of microbes. The rabbitfish group fed the medium ARA diet demonstrated a significant (p < 0.05) increase

Fig. 7. Principal component analysis (PCA) of fatty acid composition and immune function in rabbitfish fed with different levels of ARA. The figure explains correlations between variables and clustering between samples based on their scores obtained from the analysis. Circles represent samples from different dietary treatments used in the analysis, and vectors represent variables that contributed to the variance. Different fatty acids and immune parameters used in the analysis are indicated in black and blue fonts, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

in plasma lysozyme levels after one and seven months, compared to the low ARA control and the high ARA groups. Interestingly, the high ARA group (4.7% of TFA) had relatively low levels of lysozyme, which might be due to the immunosuppressive effects of excessive levels of dietary ARA [25]. In addition, to lysozyme higher complement levels were observed in both groups of fish fed the ARAsupplemented diets. Remarkably, better complement activity was observed in the moderate ARA group than in the high ARA group, which was quite similar to the trend obtained for lysozyme activity. These results substantiate previous findings obtained in Japanese sea bass (Lateolabrax japonicus) in which a diet with 1.57e4.91% ARA of TFA resulted in higher levels of both serum lysozyme and complement activity as compared to a diet with 15.96% ARA [26]. In

52

S. Nayak et al. / Fish & Shellfish Immunology 68 (2017) 46e53

addition, recent studies of dietary supplementation with ARA in carp (Ctenopharyngodon idellus) and Atlantic salmon (Salmo salar) have also reported a significant improvement in the non-specific immune function [27,28]. Serum antiprotease activities in fish are primarily due to the presence of a1-antiprotease, a2-antiplasmin and a2 macroglobulin acting as antienzymes that neutralize and inhibit invading pathogens [29]. In this study, antiprotease levels in rabbitfish plasma were marginally downregulated in the ARA-supplemented groups as compared to the unsupplemented control. However, it was previously documented that antiprotease levels are generally high in fish and are barely modulated even after immunization or infection [30], although studies have also reported enhanced antiprotease activity following application of immunostimulants and probiotics [31,32]. This suggests that dietary ARA supplementation might not be a key player in stimulating plasma antiprotease levels in rabbitfish. In addition to nonspecific immune function, rabbitfish fed with the ARA-supplemented diets had significantly higher levels of total immunoglobulins as compared to the control group, suggesting improved ability to mount adaptive immune responses. These results support the assumption that ARA-derived eicosanoid production plays a key role in the activation of humoral responses in teleosts [33]. Previous studies demonstrated that dietary ARA supplementation increases fish resistance to pathogens [34,35]. In studies carried out on Atlantic salmon, immunoglobulin levels were altered due to modification of dietary oil in the food [36,37]. Generally, low immunoglobulin levels may serve as a sign of immunodeficiency, whereas exceedingly elevated levels could suggest an underlying state of disease or infection [38]. In this study, immunoglobulin levels were moderately elevated, suggesting that the cause was not an infectious stimulus. Analysis of various immune parameters in the three experimental groups demonstrated an overall increase in innate and adaptive immune responses in fish fed the ARA supplemented diets. The optimal dietary ARA content appears to be the moderate dietary level (2.6% of TFA). Previous studies already documented that over-supplementation with LCPUFA, in general, and ARA, in particular, can cause adverse effects to the host immune system and disease resistance mechanisms [25,34]. Elevated levels of ARA, which has usually pro-inflammatory effect, might actually promote processes such as oxidative stress, resulting in the partial suppression of immune functions [25]. Apart from enhanced immune function, juvenile rabbitfish fed with the moderate ARA diet (2.6%) showed enhanced growth and a significant (p < 0.05) rate of weight gain as compared to the control and high ARA groups. Although the exact mechanism by which dietary ARA can increase growth is not clear, experiments in mammals have shown that ARA-derived PGF2 a stimulates protein synthesis, while PGE2 is involved in protein degradation [39]. Furthermore, fish supplemented with moderate ARA clearly survived better than the other two treatment groups, presumably due to enhanced resistance to the disease (Fig. 2). A large variation in mortality was evident in the 0.6% and 4.7% treatments between replicate tanks, which is not surprising, as the infection was a natural outbreak, rather than a controlled challenge. The effect of ARA supplementation on growth and survival is in line with the observed effect on immune function, showing elevated growth in fish fed ARA-supplemented diet at 2.6% of TFA, compared to both higher and lower supplementation. A similar effect of ARA on the growth of Japanese sea bass was observed by Xu et al. (2010), where fish fed with 0.46%, 1.47% and 2.59% ARA of TFA exhibited better growth rate compared to fish fed with the higher levels of 4.91%, 12.1% and 15.96% ARA of TFA. Further increase in levels of lysozyme and alternative complement were

reported to be significantly enhanced in the 2.59% and 4.91% ARA supplemented groups [26]. Similarly, guppy fry fed with a diet supplemented with ARA at 3.5% of TFA was shown to promote both survival and increased lysozyme levels [40]. To evaluate the incorporation of ARA from the diet into the rabbitfish erythrocytes, we analyzed fatty acid composition in the red blood cells. It is well documented that the fatty acid composition of circulating blood cells is affected by the nature of the dietary oils and the consumed fats [41,42]. Furthermore, dietary fatty acids are known to effect erythrocyte membrane FA composition in fish [43]. In this study, ARA-rich diets led to changes in the fatty acid composition of RBCs and increased ARA levels, suggesting that RBCs can be used to evaluate the effects of ARA uptake from supplemented diets. Along with the increase in ARA, we found a concomitant decrease in the levels of other fatty acids, including 18:1, 18:2, 18:3 and EPA, suggesting that dietary ARA replaced these FAs in RBC lipids. The decrease in the EPA/ARA ratio was predominantly due to increased ARA content. In addition, an increase in 22:4n-6 was measured in the supplemented groups, although it was only significant for the high ARA groups. This fatty acid is an elongation product of ARA. Moreover, a moderate increase in 22:5n-6 was also observed, although insignificant, suggesting desaturation of 22:4n-6 (ARA) [44]. [45e47]The PCA analysis showed a clear differentiation of samples based on their dietary ARA patterns. The ARAsupplemented samples clustered on the upper right side of the quadrant as compared to the unsupplemented group on the lower quadrant, indicating an effect of ARA supplementation on the fatty acid profile. PCA analysis demonstrated a positive association between ARA, as well as other n-6 LCPUFAs, and samples from the supplemented group, thus confirming the incorporation of ARA into erythrocytes. Furthermore, an association between all immune functions, with the exception of antiprotease, and the medium and high ARA diets was evident. Overall, this suggests that the modulation observed in immune function was influenced by dietary patterns, supporting the assumption that inclusion of dietary ARA has a modulatory effect on the immune functions of rabbitfish. In summary, addition of ARA to the diet improves rabbitfish growth, survival and immune competence. In this study, addition of 2.6% ARA to the standard fish diet provided optimal effects. Results from multivariate analyses support these observations and the impact of ARA nutrition on fatty acid and immune-related variables. Thus, ARA supplementation to rabbitfish diet could be one of several valuable nutrient supplements essential for establishing the commercial culture of this species. Acknowledgements This work was supported by the U.S. Agency for International Development, MERC Projects; M28-061(Production of rabbitfish, Siganus rivulatus, through low-impact, land based mariculture) and M33-034 (Application and scale-up of rabbitfish production in Jordan and Israel Arava/Araba valley: Phase B). The postdoctoral fellowship of SN was supported by the Israeli Council for Higher Education, PBC Program. We are grateful to Dr. Angelo Corloni for helping us seine capture wild juvenile Siganus rivulatus for the first study. References [1] F.A. Bukhari, Trials of rabbitfish siganus rivulatus production in floating cages in the red sea, Emir. J. Food Agric. 17 (2) (2005) 23e29. [2] M. Badran, M. Al-Zibdah, Environmentally friendly mariculture in coral reef areas, Mu’tah Lil-Buhoth Wad-Dirasat 18 (4) (2006) 147e169. [3] R.D. Van Anholt, F.A.T. Spanings, W.M. Koven, O. Nixon, S.E. Wendelaar Bonga, Arachidonic acid reduces the stress response of gilthead seabream Sparus

S. Nayak et al. / Fish & Shellfish Immunology 68 (2017) 46e53 aurata L, J. Exp. Biol. 207 (19) (2004) 3419e3430. [4] J.G. Bell, J.R. Sargent, Arachidonic acid in aquaculture feeds: current status and future opportunities, Aquaculture 218 (1e4) (2003) 491e499. [5] M. Bessonart, M.S. Izquierdo, M. Salhi, C.M. Hern andez-Cruz, M.M. Gonz alez, H. Fern andez-Palacios, Effect of dietary arachidonic acid levels on growth and survival of gilthead sea bream (Sparus aurata L.) larvae, Aquaculture 179 (1e4) (1999) 265e275. [6] W. Koven, Y. Barr, S. Lutzky, I. Ben-Atia, R. Weiss, M. Harel, P. Behrens, A. Tandler, The effect of dietary arachidonic acid (20:4n6) on growth, survival and resistance to handling stress in gilthead seabream (Sparus aurata) larvae, Aquaculture 193 (1e2) (2001) 107e122. [7] E. Ricciotti, G.A. Fitzgerald, Prostaglandins and inflammation, Arteriosclerosis, Thrombosis, Vasc. Biol. 31 (5) (2011) 986e1000. [8] E.R. Greene, S. Huang, C.N. Serhan, D. Panigrahy, Regulation of inflammation in cancer by eicosanoids, Prostagl. Other Lipid Mediat. 96 (1e4) (2011) 27e36.  [9] M.A. De Pablo Martinez, G. Alvarez De Cienfuegos, Modulatory effects of dietary lipids on immune system functions, Immunol. Cell Biol. 78 (1) (2000) 31e39. [10] F.-H. Wang, M.-Q. Xie, A.-X. Li, A novel protein isolated from the serum of rabbitfish (Siganus oramin) is lethal to Cryptocaryon irritans, Fish Shellfish Immunol. 29 (1) (2010) 32e41. [11] R. Li, A. Li, Antibacterial efficacy of recombinant Siganus oramin l-amino acid oxidase expressed in Pichia pastoris, Fish Shellfish Immunol. 41 (2) (2014) 356e361. [12] A.S.O. Lie, E. Froysadal, Ø. Evensen Lie, SR, A. Schultz, O.S. Jacobsen, J. Aamand Sørensen, O. Evensen, A. Smed-Sorensen, Study on lysozyme activity in some fish species, Dis. Aquatic Org. 6 (1989) 1e5. [13] B. Magnadottir, Immunological control of fish diseases, Mar. Biotechnol. 12 (4) (2010) 361e379. [14] G. Sharon, P.R. Nath, N. Isakov, D. Zilberg, Evaluation of guppy (Poecilia reticulata Peters) immunization against Tetrahymena sp. by enzyme-linked immunosorbent assay (ELISA), Veterinary Parasitol. 205 (1e2) (2014) 28e37. [15] M. Izquierdo, W. Koven, Larval Fish Nutrition, Wiley-Blackwell, John Wiley & Sons, Inc., U.K., 2011. [16] M.J. Salini, N.M. Wade, B.C. Araújo, G.M. Turchini, B.D. Glencross, Eicosapentaenoic acid, arachidonic acid and eicosanoid metabolism in juvenile barramundi Lates calcarifer, Lipids 51 (8) (2016) 973e988. [17] D. Montero, G. Terova, S. Rimoldi, M.B. Betancor, E. Atalah, S. Torrecillas, M.J. Caballero, M.J. Zamorano, M. Izquierdo, Modulation of the expression of components of the stress response by dietary arachidonic acid in european sea bass (Dicentrarchus labrax) larvae, Lipids 50 (10) (2015) 1029e1041. [18] M. Furne, E. Holen, P. Araujo, K.K. Lie, M. Moren, Cytokine gene expression and prostaglandin production in head kidney leukocytes isolated from Atlantic cod (Gadus morhua) added different levels of arachidonic acid and eicosapentaenoic acid, Fish Shellfish Immunol. 34 (3) (2013) 770e777. [19] J.J. Tian, C.X. Lei, H. Ji, A. Jin, Role of cyclooxygenase-mediated metabolites in lipid metabolism and expression of some immune-related genes in juvenile grass carp (Ctenopharyngodon idellus) fed arachidonic acid, Fish Physiology Biochem. 43 (3) (2017) 703e717. [20] R.D. Van Anholt, F.A.T. Spanings, W.M. Koven, S.E. Wendelaar Bonga, Dietary supplementation with arachidonic acid in tilapia (Oreochromis mossambicus) reveals physiological effects not mediated by prostaglandins, General Comp. Endocrinol. 139 (3) (2004) 215e226. [21] R. Fischer, A. Konkel, H. Mehling, K. Blossey, A. Gapelyuk, N. Wessel, C. Von Schacky, R. Dechend, D.N. Muller, M. Rothe, F.C. Luft, K. Weylandt, W.H. Schunck, Dietary omega-3 fatty acids modulate the eicosanoid profi le in man primarily via the CYP-epoxygenase pathway, J. Lipid Res. 55 (6) (2014) 1150e1164. [22] P.C. Calder, Marine omega-3 fatty acids and inflammatory processes: effects, mechanisms and clinical relevance, Biochimica Biophysica Acta - Mol. Cell Biol. Lipids 1851 (4) (2015) 469e484. [23] A.F. Rowley, J. Knight, P. Lloyd-Evans, J.W. Holland, P.J. Vickers, Eicosanoids and their role in immune modulation in fish-a brief overview, Fish Shellfish Immunol. 5 (8) (1995) 549e567. [24] W.B. Van Muiswinkel, M. Nakao, A short history of research on immunity to infectious diseases in fish, Dev. Comp. Immunol. 43 (2) (2014) 130e150. [25] J.E. Kinsella, B. Lokesh, S. Broughton, J. Whelan, Dietary polyunsaturated fatty acids and eicosanoids: potential effects on the modulation of inflammatory and immune cells: an overview, Nutrition 6 (1) (1990) 24e44. [26] H. Xu, Q. Ai, K. Mai, W. Xu, J. Wang, H. Ma, W. Zhang, X. Wang, Z. Liufu, Effects of dietary arachidonic acid on growth performance, survival, immune response and tissue fatty acid composition of juvenile Japanese seabass,

53

Lateolabrax japonicus, Aquaculture 307 (1e2) (2010) 75e82. [27] J. Tian, H. Ji, H. Oku, J. Zhou, Effects of dietary arachidonic acid (ARA) on lipid metabolism and health status of juvenile grass carp, Ctenopharyngodon idellus, Aquaculture 430 (2014) 57e65. ndez, [28] P. Dantagnan, K. Gonzalez, M. Hevia, M.B. Betancor, A.J. Herna A. Borquez, D. Montero, Effect of the arachidonic acid/vitamin E interaction on the immune response of juvenile Atlantic salmon (Salmo salar) challenged against Piscirickettsia salmonis, Aquac. Nutr. (2016), http://dx.doi.org/ 10.1111/anu.12438. [29] A.E. Ellis, Innate host defense mechanisms of fish against viruses and bacteria, Dev. Comp. Immunol. 25 (8e9) (2001) 827e839. [30] S.K. Nayak, Probiotics and immunity: a fish perspective, Fish Shellfish Immunol. 29 (1) (2010) 2e14. [31] L.T. Guzm an-Villanueva, D. Tovar-Ramírez, E. Gisbert, H. Cordero, F.A. Guardiola, A. Cuesta, J. Meseguer, F. Ascencio-Valle, M.A. Esteban, Dietary administration of b-1,3/1,6-glucan and probiotic strain Shewanella putrefaciens, single or combined, on gilthead seabream growth, immune responses and gene expression, Fish Shellfish Immunol. 39 (1) (2014) 34e41. [32] J. Brunt, A. Newaj-Fyzul, B. Austin, The development of probiotics for the control of multiple bacterial diseases of rainbow trout, Oncorhynchus mykiss (Walbaum), J. Fish Dis. 30 (10) (2007) 573e579. [33] M.J. James, R.A. Gibson, L.G. Cleland, Dietary polyunsaturated fatty acids and inflammatory mediator production, Am. J. Clin. Nutr. 71 (1 SUPPL) (2000). [34] I. Khozin-Goldberg, Z. Cohen, M. Pimenta-Leibowitz, J. Nechev, D. Zilberg, Feeding with arachidonic acid-rich triacylglycerols from the microalga Parietochloris incisa improved recovery of guppies from infection with Tetrahymena sp, Aquaculture 255 (1e4) (2006) 142e150. [35] K.D. Thompson, M.F. Tatner, R.J. Henderson, Effects of dietary (n-3) and (n-6) polyunsaturated fatty acid ratio on the immune response of Atlantic salmon, Salmo salar L, Aquac. Nutr. 2 (1) (1996) 21e31. [36] J.G. Bell, I. Ashton, C.J. Secombes, B.R. Weitzel, J.R. Dick, J.R. Sargent, Dietary lipid affects phospholipid fatty acid compositions, eicosanoid production and immune function in Atlantic salmon (Salmo salar), Prostagl. Leukot. Essent. Fat. Acids 54 (3) (1996) 173e182. [37] L. Martinez-Rubio, T. Evensen, A. Krasnov, S.M. Jørgensen, S. Wadsworth, K. Ruohonen, J.L.G. Vecino, D.R. Tocher, Effects of functional feeds on the lipid composition, transcriptomic responses and pathology in heart of Atlantic salmon (Salmo salar L.) before and after experimental challenge with Piscine Myocarditis Virus (PMCV), BMC Genomics 15 (1) (2014). [38] B. Magnadottir, B.K. Gudmundsdottir, A comparison of total and specific immunoglobulin levels in healthy Atlantic salmon (Salmo salar L.) and in salmon naturally infected with Aeromonas salmonicida subsp. achromogenes, Veterinary Immunology and Immunopathology 32 (1) (1992) 179e189. [39] R.M. Palmer, Prostaglandins and the control of muscle protein synthesis and degradation, Prostaglandins, Leukot. Essent. Fat. Acids 39 (2) (1990) 95e104. [40] P.R. Nath, I. Khozin-Goldberg, Z. Cohen, S. Boussiba, D. Zilberg, Dietary supplementation with the microalgae Parietochloris incisa increases survival and stress resistance in guppy (Poecilia reticulata) fry, Aquac. Nutr. 18 (2) (2012) 167e180. [41] R.C. Block, W.S. Harris, J.V. Pottala, Determinants of blood cell Omega-3 fatty acid content, open biomarkers J. 1 (2008) 1e6. [42] S.K. Raatz, D. Bibus, W. Thomas, P. Kris-Etherton, Human nutrition and metabolism: total fat intake modifies plasma fatty acid composition in humans, J. Nutr. 131 (2) (2001) 231e234. [43] W.-K. Ng, P.-K. Lim, H. Sidek, The influence of a dietary lipid source on growth, muscle fatty acid composition and erythrocyte osmotic fragility of hybrid tilapia, Fish Physiology Biochem. 25 (4) (2001) 301e310. [44] F. Linares, R.J. Henderson, Incorporation of 14C-labelled polyunsaturated fatty acids by juvenile turbot, Scophthalmus maximus (L.) in vivo, J. Fish Biol. 38 (3) (1991) 335e347. [45] G. Rosenlund, Effect of alternative lipid sources on long-term growth performance and quality of Atlantic salmon (Salmo salar L.), Aquac. Res. 32 (SUPPL. 1) (2001) 323e328. [46] B.E. Torstensen, L. Frøyland, Ø. Lie, Replacing dietary fish oil with increasing levels of rapeseed oil and olive oil - effects on Atlantic salmon (Salmo salar L.) tissue and lipoprotein lipid composition and lipogenic enzyme activities, Aquac. Nutr. 10 (3) (2004) 175e192. [47] R.E. Olsen, G.L. Taranger, T. Svåsand, O.T. Skilbrei, Improved method for triacylglycerol-derived fatty acid profiling by various non-lethal and lethal sampling techniques in Atlantic salmon, Aquac. Environ. Interact. 4 (3) (2013) 251e261.