Functional diversification of teleost Fads2 fatty acyl desaturases occurs independently of the trophic level

Abstract

The long-chain (≥C₂₀) polyunsaturated fatty acid biosynthesis capacity of fish varies among species, with trophic level hypothesised as a major factor. The biosynthesis capacity is largely dependent upon the presence of functionally diversified fatty acyl desaturase 2 (Fads2) enzymes, since many teleosts have lost the gene encoding a Δ5 desaturase (Fads1). The present study aimed to characterise Fads2 from four teleosts occupying different trophic levels, namely Sarpa salpa, Chelon labrosus, Pegusa lascaris and Atherina presbyter, which were selected based on available data on functions of Fads2 from closely related species. Therefore, we had insight into the variability of Fads2 within the same phylogenetic group. Our results showed that Fads2 from S. salpa and C. labrosus were both Δ6 desaturases with further Δ8 activity while P. lascaris and A. presbyter Fads2 showed Δ4 activity. Fads2 activities of herbivorous S. salpa are consistent with those reported for carnivorous Sparidae species. The results suggested that trophic level might not directly drive diversification of teleost Fads2 as initially hypothesised, and other factors such as the species' phylogeny appeared to be more influential. In agreement, Fads2 activities from P. lascaris and A. presbyter were similar to their corresponding phylogenetic counterparts Solea senegalensis and Chirostoma estor.

Figure 1

Biosynthetic pathways of LC-PUFA from the precursors linoleic acid (18:2n-6) and α-linolenic acid (18:3n-3) in teleosts. Black and grey arrows show reactions catalysed by fatty acyl desaturase (“Δx”) and fatty acyl elongase (“Elovl”) activities, respectively.

Figure 2

Alignment of the deduced amino acid (aa) sequences of the novel fatty acyl desaturases of Sarpa salpa, Chelon labrosus, Pegusa lascaris and Atherina presbyter. Identical residues are shaded black, heme binding motif is indicated by asterisks, histidine boxes are dashed black framed, and regions determining regioselectivity is solid black framed.

Figure 3

Phylogenetic tree including the deduced amino acid (aa) sequences of the fatty acyl desaturases of Sarpa salpa, Chelon labrosus, Pegusa lascaris and Atherina presbyter. The horizontal branch length is proportional to the aa substitution rate per site. Demonstrated desaturase activities are included in all Fads-like sequences as “Δx”. The asterisk (“*”) in some Δ6 desaturases denotes in demonstrated ability of these enzymes to operate towards both C₁₈ (e.g. 18:3n-3) and C₂₄ (e.g. 24:5n-3) substrates.

Figure 4

Distribution of fads2 mRNA levels in tissues of Sarpa salpa, Chelon labrosus, Pegusa lascaris and Atherina presbyter. Data are shown as geometric mean log normalised expression ratios ± standard errors (n = 4, except for heart and muscle of S. salpa where n = 3, brain of C. labrosus where n = 3, and brain and heart of A. presbyter where n = 2 and n = 1, respectively). Different letters denote significant differences among tissue within each species (P < 0.05, One-way ANOVA, Tukey HSD test).