Fatty acid transfer in the food web of a coastal Mediterranean lagoon: Evidence for high arachidonic acid retention in fish

Abstract

The transfer of fatty acids (FAs) in the food web of a Mediterranean lagoon was studied using FA compositional patterns across several trophic levels. The structure of the food web was inferred from C and N stable isotopes values and an isotope mixing model was used in order to estimate the relative contribution of the different potential food sources to the biomass of consumers. Bidimensional plots of FA composition of food web components against their δ¹⁵N values indicated a general trend of increasing proportions of highly unsaturated fatty acids (HUFAs) with increasing trophic levels while the proportions of saturated fatty acids (SAFAs) and 18-carbon polyunsaturated fatty acids (PUFAs) decreased. Using the relative contributions of food sources to consumers and their FA compositions, a model was built in order to estimate the PUFA composition of consumer mixed diets which was compared to consumer PUFA profiles. The latter allowed the identification of the PUFAs which were mostly enriched/retained in consumer lipids. There was a surprisingly high retention of arachidonic acid (ARA), a trend which challenges the idea of low ARA needs in marine fish and suggests the important physiological role of this essential FA for fish in estuarine environments.

Introduction

Mediterranean lagoons are, as with other coastal areas, very productive ecosystems that play a crucial role as feeding grounds for adult and juvenile fish and invertebrate species (Kjerfve, 1994, Razinkovas et al., 2008). However, in contrast to other coastal ecosystems (e.g. estuaries, fjords), Mediterranean lagoons are, for climatic reasons, deprived of significant freshwater inputs (Elliott et al., 2002) and often their communication with adjacent coastal areas is limited (Franco et al., 2008). As a consequence, consumers in Mediterranean lagoons food webs are expected to rely on a mixture of autochthonous particulate detrital organic material, bacteria, microalgae, heterotrophic protists and fungi (Elliott et al., 2002, Franco et al., 2008, Vizzini and Mazzola, 2008). In such detritus-dominated environments, basal organic matter sources (e.g. suspended and sediment particulate organic matter) are characterized by high levels of saturated fatty acids (SAFAs) and 18-carbon polyunsaturated FAs (≥ 2 double bonds; PUFAs) and relatively low levels of highly unsaturated FAs (≥ 20 carbon atoms and ≥3 double bonds; HUFAs) (Mudge et al., 1998; Alfaro et al., 2006, Richoux and Froneman, 2008; Copeman et al., 2009; ).

In aquatic ecosystems, n-3 and n-6 polyunsaturated FAs (PUFAs) have important effects on growth, reproduction and survival of invertebrates and fish (Sargent et al., 1999a, Arts et al., 2001). Among these molecules, the three major HUFAs: docosahexaenoic acid (22:6n-3/DHA) eicosapentaenoic acid (20:5n-3/EPA) and arachidonic acid (20:4n-6/ARA) are particularly important. For example, DHA is thought to be at least implicated in several cell membrane properties such as permeability, membrane fusion or elasticity (Arts and Kohler, 2009). EPA and ARA are precursors for different molecular families of the eicosanoid class of hormones, which are locally-active signalling molecules implicated in inflammation, immunity, energy allocation, mineral balance and reproductive success in animals (Sargent et al., 1999a; Schmitz and Ecker, 2008; Parrish, 2009). ARA-derived eicosanoids are more biologically active than those derived from EPA but EPA competitively inhibits the formation of eicosanoids derived from ARA (Sargent et al., 1999a, Schmitz and Ecker, 2008). Hence, the EPA:ARA ratio in tissues determines the action of eicosanoid on fish physiology.

n-3 and n-6 PUFAs are generally formed in primary producers and certain aquatic fungi while most marine metazoa cannot synthesize them efficiently (Olsen, 1998, Parrish, 2009). These compounds are then circulated and concentrated through the food web (Olsen, 1998, Parrish, 2009). Therefore, as most marine animals (fish, invertebrates) rely on dietary inputs for the supply of 18:3n-3, 18:2n-6, ARA, EPA and DHA, they are considered as essential fatty acids: (EFAs) (Arts et al., 2001). As the incorporation of EFAs in animal tissues is based on low specificity acylases and trans-acylases, tissue composition and therefore membrane structure and eicosanoid regulation are partially influenced by the relative proportions of these compounds in the environment (Copeman and Parrish, 2003). For example, 18-carbon PUFAs which are abundant in macrophyte (e.g. macroalgae, immerged halophytes) detritus can compete for the enzymes that esterify ARA, EPA and DHA in consumer tissues. Thus, excessive dietary proportions of these compounds relative to HUFA proportions could have negative effects on fish growth and survival (Sargent et al., 1999a). Hence, EFA relative availability in aquatic food webs can affect intrinsic health, growth and reproduction of animal populations (Olsen, 1998; Ahlgren et al., 2005).

Most marine fish store their lipids in the form of neutral lipids (NLs), whereas cell-membrane lipids are mostly in the form of polar lipids (PLs) (Dalsgaard et al., 2003). NLs are storage lipids and are considered to better reflect diet as their FA composition largely depends on the most abundant dietary FAs (Dalsgaard et al., 2003, Lands, 2009). Alternatively, PLs are the major constituent of cell membranes and, because of their physiological role, their FA composition is determined by more selective processes, regulated to a certain extent by specific bodily needs and functions (Ackman, 1998, Lands, 2009). Indeed, PUFA retention patterns in fish tissues have been shown to respond to parameters such as salinity, temperature (Cordier et al., 2002) or dietary lipid FA composition (Copeman et al., 2002, Fountoulaki et al., 2003). Therefore, as PLs tend to show a certain degree of homeostasis, and in the case that the proportion of a given PUFA in dietary lipids is lower relative to consumer requirements, one could expect a selective enrichment of these compounds from the food into consumer lipids (Copeman et al., 2002, Castell et al., 2003, Fountoulaki et al., 2003, Hessen and Leu, 2006). Similarly, PUFAs in excess should occur in lower proportions in consumers than in their food.

In the present study, our objectives where (1) to investigate the changes of FA patterns across multiple trophic levels, in a Mediterranean coastal lagoon and (2) to identify which PUFAs are mostly accumulated in lipids in the lagoon’s principal fish species compared to their food. Consequently, carbon and nitrogen stable isotope analyses and an isotope mixing model were used in order to estimate the relative contribution of the different potential food sources to the biomass of fish. Then, using food source relative contributions and their PUFA compositions, a model was built in order to estimate the PUFA composition of consumer diets which was compared to consumer FA profiles.