Skin Mucus Fatty Acid Composition of Gilthead Sea Bream ( Sparus Aurata ) : A Descriptive Study in Fish Fed Low and High Fish Meal Diets

Terrestrial protein and lipid sources are commonly used as substitutes for marine fishery-derived raw ingredients in fish diets. However, their use is related with several side-effects on marine fish performance, health, or disease resistance. Physical barriers of the skin, gills, and gut constitute the primary defense mechanism of fish. Skin mucus mucosal mucins, water, proteins, ions, and lipids determine the physical, chemical, and protective characteristics of skin mucus. Very little is known about the influence of diet composition on fish skin mucus fatty acid profile. Gilthead sea bream skin mucus contained 10% of total lipids (TL), which consisted of 50–60% neutral (NL) and 40–50% polar lipids (PL) fractions. Σn−3 long chain polyunsaturated fatty acids (LC-PUFA) deposition was preferential in the NL fraction, whereas Σn−6LC-PUFA accumulation was similar in both lipid classes. Docosahexaenoic acid (DHA; 22:6n−3) was the main LC-PUFA stored in skin mucus (14% TL) in relation to eicosapentaenoic acid (EPA; 20:5n−3) (2–3% TL) and arachidonic acid (ARA; 20:4n−6) (2% TL). This study denotes the importance of DHA as component of skin mucus lipids compared to other essential fatty acids, such as EPA and ARA, as well as importance of maintaining an adequate Σn−3/ Σn−6 ratio, regardless of dietary intake.


Introduction
Terrestrial protein and lipid sources are commonly used as substitutes to marine fishery-derived proteins and lipids in gilthead sea bream (Sparus aurata) commercial diets [1][2][3].However, even when diets are formulated to cover the theoretical nutrient requirements for essential fatty acids or amino acids [4], variations in their dietary percentage, and thus in specific ratios among them, have been related with several side-effects on marine fish growth performance, health, welfare, or disease resistance.Furthermore, high levels of dietary plant protein sources may negatively affect the bioavailability of micronutrients [5], contain endogenous anti-nutritional factors, and alter fatty acid profiles, affecting not only fish growth but also fish health and disease resistance [6][7][8][9].
Physical barriers of the skin, gills, and gut constitute the primary defense mechanism of fish exposed to an external infectious or damaging agent [10].Fish skin shares common components with mammalian mucosal surfaces, such as the presence of macrophages, lymphocytes, eosinophilic granulocytes, dendritic cells, an organized cytokine response, and scattered mucus-secreting cells in the epithelium [11], particularly in the stratum spinosum.Skin mucus acts as a physical, semipermeable, dynamic, chemical, and biological barrier [12,13].The physical, chemical, and protective characteristics of mucus are determined by its composition, which is mainly based on water, mucosal mucins, and a mixture of proteins, ions, and lipids [11].The influence of water bacterial load or parasite presence on the composition and physicochemical properties of skin mucus has been described through recognition of pathogen-associated molecular patterns (PAMPs) [14][15][16].In higher vertebrates, it has been demonstrated that skin lipid film can be modulated through the diet more efficiently than, for example, through direct topical application [17].However, in fish, very little is known about the effects of high dietary levels of plant meals and oils on the composition and particularly on the fatty acid profiles of skin mucus-this represents the main objective of this study for gilthead sea bream (Sparus aurata), one of the most cultivated species in the Mediterranean.

Discussion
The physical, chemical, and protective characteristics of fish skin mucus are determined by its composition [11], however, its lipid composition and possible functions have been scarcely studied [18].Mucus composition studies from other mucosal surfaces, such as the gut, indicate that mucus lipids are good biomarker candidates and that variations in its composition or concentrations may indicate its functional role [18,19].In the present study, and regardless of diet composition, gilthead sea bream skin mucus lipid content (10-11% dw: 0.8-1% ww) was higher than that reported in previous studies for several fresh water fish species, such as African catfish (Clarias gariepinus; 5.9% dw) and Malaysia catfish (Clarias sp.1; 4.0% dw) [20], or marine species such as Seriola dumerli (0.62% ww) [21] or Verasper variegatus (0.66% ww) [21].Despite the different dietary intake, the gilthead skin mucus NL fraction accounted for approximately 6.1-6.2%(dw) of the total lipid content in fish fed both diets, whereas the PL fraction represented 40% (4.2% dw) in fish fed L-FM-based diets and 47% (5.3% dw) in fish fed H-FM-based diets.These percentages represent a 4-and 2-fold increase in PL content in relation to dietary contents, respectively, and emphasize the importance of PL in fish skin lipid mucus composition.In higher vertebrates, the orientation of the PL lipophilic region, together with the nature of the fatty acids, determines the hydrophobicity of the mucus gel layer, forming a "non-wettable" resistant layer towards the external surface of fatty acid tails [22][23][24].In fact, in reference [18], the authors established an interaction between the fish mucus PL and glycoprotein fractions in relation to the mucus layer viscosity, with fish mucus viscosity increasing with higher levels of mucus PL.In this sense, for example, it has been suggested that Labroides dimidiatus skin mucus PL content influences the protective role and the rigidity of the mucus layer in relation to its ability to survive in a parasitic environment [25].
The total skin mucus lipid fatty acid profiles of fish fed both dietary treatments were similar regardless of the different dietary profile.The similar percentages observed in both groups for saturated, Σn−3, Σn−3LC-PUFA, and Σn−6LC-PUFA fractions denote the essentiality of these groups of fatty acids in gilthead sea bream mucus.Particularly, the high percentages of palmitic and stearic fatty acids present in gilthead seabream skin mucus are in accordance with previous findings in other fish species, such as Seriola dumerli, V. variegatus [21], L. dimidiatus [25], and Channa striatus [26], and they are probably related to its structural role in the sn1 position of PL.Unfortunately, there is no information available about the fatty acid composition of fish mucus PL.However, based on studies on higher vertebrates, phosphatidylcholine and lysophosphatidylcholine are the major species in colonic mucus [24,27], which contain typically one saturated (palmitic or stearic acids) and one unsaturated fatty acid (oleic/linoleic acids) [24].Indeed, in the present study, palmitic and stearic fatty acid contents were specifically increased in the skin mucus PL fraction.
Furthermore, the selective incorporation of Σn−3LC-PUFA and Σn−6LC-PUFA fractions observed in the total lipid skin mucus of gilthead seabream is remarkable when fed a L-FM-based diet, however they differ in the fraction where they were incorporated.The Σn−3LC-PUFA fraction was preferably incorporated in the NL fraction, mainly as a consequence of selective deposition of DHA and DPA, whereas the Σn−6LC-PUFA fraction was indistinctly incorporated in both lipid fractions, mainly due to a higher retention of 20:3n−6, ARA, 22:4n−6, and 22:5n−6.The increase in Σn−3LC-PUFA percentages in the NL fraction suggests not only their importance as NL components in gilthead skin mucus, but also makes evident the incapacity of mucus and associated biota to use them as energy substrates, as described in fish tissues [28] and supported by the percentages of oleic acid detected in both lipid fractions.Additionally, a trend to higher DHA and ARA and a lower EPA deposition in the skin mucus PL fraction, regardless of the diet, may be indicative of changes in the PL species' percentages in relation to the physical and immune properties of mucus.Gilthead seabream skin mucus had a relatively high percentage of DHA (14% TL) in relation to EPA (2-3% TL) and ARA (2% TL), in agreement with previous results in other marine fish species, such as Seriola dumerli [21], which may be related with a higher amount of phosphatidylethanolamine and phosphatidylserine where it is preferentially deposited.Contrary to the present findings, in other freshwater fish species, such as haruan, the percentages of ARA in fish mucus represented about 20X DHA content, and this has been related with facilitating wound healing in this fish species [26] due to its cellular aggregation and blood clotting capacities, among others.
In conclusion, our results emphasize the importance of the PL fraction in gilthead sea bream skin mucus, irrespective of its dietary intake.Additionally, the results obtained show the importance of LC-PUFA fractions in NL and PL lipid fractions based mainly on the selective deposition found in fish fed lower LC-PUFA percentages.In particular, Σn−3LC-PUFA (DHA and DPA) deposition has been found to be preferential in the NL fraction, whereas Σn−6LC-PUFA (20:3n−6, ARA, 22:4n−6, and 22:5n−6) deposition appears to be similar in both lipid classes.Due to the importance of the fish mucus skin layer in an aqueous immediate environment, further research is needed in order to understand the effects of feed composition on its protective function and how it can be modified through diet or diet additives, in terms of potentiating its protective functions.

Materials and Methods
Two diets were formulated to contain two levels of fish meal (FM) as follows: a high-FM-based diet (H-FM; 63% FM) and a low-FM-based diet (L-FM; 15% FM).Diets were isoenergetic and isonitrogenous and were formulated to meet all known nutritional requirements for several marine finfish [4] (Table 2).Diets were produced by BioMar (BioMar Tech-Centre, Brande, Denmark).Diet composition, proximate analysis, lipid class composition, and fatty acid profiles are shown in Tables 2 and 3, respectively.The feeding trial was carried out at the experimental facilities of the IU-ECOAQUA of the Universidad de Las Palmas de Gran Canaria (Las Palmas, Canary Islands, Spain).Gilthead sea bream juveniles, obtained from natural spawning from our own broodstock (IU-ECOAQUA), with an average initial weight and length (mean ± SD) of 22.5 ± 1.5 g and 11.7 ± 0.4 cm, respectively, were fed manually until apparent satiation with one of the two experimental diets for 36 weeks (3 times a day, 6 days a week).Tanks were in a flow-through system with filtered seawater at a natural photoperiod (12L:12D).Water-dissolved oxygen and temperature ranged between 6.5-6.9 ppm and 20.8-24.3• C, respectively.After the feeding period, skin mucus samples were collected from fish by scraping the dorso-lateral surface of gilthead seabream using a spatula, taking care to avoid contamination with scales.In order to obtain enough skin mucus to perform the lipid analyses, equal samples of mucus were pooled into two pools by diet (2 pools of 9 fish by diet; 3 fish/tank; 3 tanks/diet).Crude lipid was extracted following the method described in [29].One of the skin mucus pools was used for quantification of total lipid fatty acid methyl esters (FAMEs), whereas the second one was used for quantification of neutral and polar fractions of the total lipids, followed by FAMEs quantification.Neutral and polar fractions of the total lipids were separated by adsorption chromatography on silica cartridges Sep-Pak®Classic (Waters, Ireland), using chloroform and chloroform/methanol (49:1; v:v) as solvent for the neutral lipid fraction, followed by the elution of the polar fraction with methanol according to [30].Fatty acid methyl esters were obtained by transmethylation with 1% sulphuric acid in methanol, as described in [31], and separated by gas chromatography (GC-14A, Shimadzu, Japan) in a Supercolovax-10-fused silica capillary column (Supelco, Bellefonte, USA) using helium as carrier gas, following the conditions described in [32].Fatty acid methyl esters were quantified by a flame ionizator detector and identified by comparison with external and well-characterized fish oil standards (EPA 28, Nippai, Ltd Tokyo, Japan).Fish final weight data was tested for normality and homogeneity of variance and a Student's t-test was used to determine significant differences between fish fed the

Table 1 .
Lipid content and fatty acid composition (% of total identified fatty acids) of gilthead sea bream (Sparus aurata) skin mucus total lipids, neutral, and polar fractions.

Table 3 .
Lipid content and fatty acid composition (% of total identified fatty acids) of the experimental diets' total lipids, neutral, and polar fractions.