Can Artemia franciscana produce essential fatty acids? Unveiling the capacity of brine shrimp to biosynthesise long-chain polyunsaturated fatty acids

Aquaculture 563 (2023) 738869 Available online 29 September 2022 0044-8486/© 2022 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/bync/4.0/). Can Artemia franciscana produce essential fatty acids? Unveiling the capacity of brine shrimp to biosynthesise long-chain polyunsaturated fatty acids Marc Ramos-Llorens a , Alberto Ribes-Navarro a , Juan C. Navarro a , Francisco Hontoria a , Naoki Kabeya b , ´ Oscar Monroig a , * a Instituto de Acuicultura de Torre de la Sal (IATS), CSIC, 12595 Ribera de Cabanes, Castell´ on, Spain b Department of Marine Biosciences, Tokyo University of Marine Science and Technology, Konan 4-5-7, Minato, Tokyo, Japan ARTICLE INFO Keywords: Artemia franciscana Elongation of very long-chain fatty acid protein Essential fatty acids Fatty acyl desaturase LC-PUFA biosynthesis ABSTRACT Artemia nauplii are widely used as live preys for feeding early life-cycle stages of marine finfish and shrimp. However, a major drawback associated to using Artemia is their deficient nutritional value for marine larvae that is primarily linked to suboptimal levels of long-chain polyunsaturated fatty acids (LC-PUFA), essential nutrients that guarantee normal growth and development of animals. While common marine hatchery procedures involve enrichment processes to address such nutritional deficiencies, the specific drivers accounting for the naturally occurring variability in LC-PUFA of Artemia are largely unknown. Biosynthesis, along diet, is one of the main factors determining the LC-PUFA profiles in animals and it depends upon the repertoire and function of fatty acyl elongases and desaturases existing in a particular species. The aim of this study was the molecular and functional characterisation of all elongase and desaturase enzymes involved in the LC-PUFA biosynthesis from Artemia franciscana, arguably the most commonly used Artemia species. Seven out of eight elongases (termed “Elo1-8”) found in A. franciscana had distinctive features of LC-PUFA biosynthesising elongases. Consistently, functional assays showed these elongases were able to elongate multiple substrates and thus enabling A. franciscana to perform all elongation reactions of the LC-PUFA biosynthetic pathways. While all seven functionally characterised elongases from A. franciscana showed activity towards multiple substrates including C 18 to C 22 polyunsaturated fatty acids, particularly high activity was detected for the Elovl8 orthologue (termed herein as “Elo7”). Additionally, A. franciscana was found to have three genes of the so-called “first” desaturases (“Des1-3”), and lack other key LC-PUFA biosynthesising desaturases such as methyl-end and front-end desaturases. Two desaturases have the expected Δ9 desaturase activity, whereas a third one showed Δ12 activity. Neither methylend nor front-end desaturases were found in A. franciscana. In conclusion, this study demonstrated that A. franciscana has high elongation capacity but its overall LC-PUFA biosynthesis can be regarded as limited due to the lack of an adequate complement of fatty acyl desaturases in its genome. 1. Introduction Artemia, particularly their early life-cycle stages (nauplii), are arguably the most commonly used live prey in marine larviculture due to being cost-effective, easy-to-handle and polyvalent (Sorgeloos et al., 2001). However, a major drawback associated to using Artemia nauplii as feed is their deficient nutritional value for marine finfish larvae that is primarily linked to suboptimal levels of long-chain polyunsaturated fatty acids (LC-PUFA). Indeed, LC-PUFA, including eicosapentaenoic acid (EPA, 20:5n-3), arachidonic acid (ARA, 20:4n-6) and docosahexaenoic acid (DHA, 22:6n-3), are essential nutrients that support optimal growth and survival of vertebrates including fish (Sargent et al., 1999; Ruxton et al., 2007; Swanson et al., 2012; Calder, 2018). While some fish species have the ability to biosynthesise LC-PUFA (Castro et al., 2016), marine carnivorous fish including breams, basses, tunas, amberjacks, etc., are largely regarded as having limited capacity for endogenous production of LC-PUFA (Xie et al., 2021; Monroig et al., 2022). Consequently, diets for marine finfish must contain adequate levels of LC- * Corresponding author. E-mail address: [email protected] (´ O. Monroig). Contents lists available at ScienceDirect Aquaculture journal homepage: www.elsevier.com/locate/aquaculture https://doi.org/10.1016/j.aquaculture.2022.738869 Received 30 May 2022; Received in revised form 23 August 2022; Accepted 25 September 2022 Aquaculture 563 (2023) 738869 2 PUFA to meet the requirements for these essential nutrients (NRC, 2011), with such demands being particularly high during early development where neural tissues accumulating these compounds are rapidly forming (Uauy et al., 2001; Janssen and Kiliaan, 2014). The fatty acid (FA) profiles of newly hatched Artemia nauplii are typically characterised by having some levels of EPA, but trace or non-detectable levels of DHA (Navarro et al., 1993). However, levels of EPA are largely variable among cysts and nauplii from different geographical origin and this led to classify Artemia into two main groups, namely “marine-type Artemia” with relatively high levels of EPA and low α -linolenic acid (ALA), and “freshwater-type Artemia” with relatively low levels of EPA and high ALA (Watanabe et al., 1978; Watanabe et al., 1980; Navarro et al., 1993). While this classification has been widely used, the specific drivers accounting for the difference in the FA profiles of Artemia remain largely unknown. The FA profiles of animal tissues are largely determined by two major factors, namely diet and endogenous production (biosynthesis). Previous studies have suggested that FA composition of food available in biotopes inhabited by Artemia are responsible for the abovementioned variability of LC-PUFA reported in Artemia cysts (Ruiz et al., 2007). For naupliar stages, it has been commonly reported that the FA composition of enrichment diets used to enhance the nutritional value of Artemia as fish and shrimp diet clearly impact the FA profiles of nauplii (Coutteau and Sorgeloos, 1997). Despite this process is known as “bioencapsulation”, there is evidence suggesting that Artemia nauplii are not just mere inert “capsules”. Some examples of lipid metabolic processes occurring within Artemia when subject to LC-PUFA enrichment processes include retroconversion of DHA into EPA (Navarro et al., 1999), selective LC-PUFA esterification patterns (Ando et al., 1997; Reis et al., 2017), and remodelling of dietary lipid classes (Guinot et al., 2013a, 2013b). As noted above, biosynthesis can also account for abundance of LC-PUFA in lipids from animals and, strikingly, such metabolic capacity is poorly understood in Artemia despite being regarded as a primary food source of these compounds in the culture of early developmental stages of marine larvae. The biosynthesis of LC-PUFA in animals is determined by the repertoire and function of certain gene families commonly known as fatty acyl elongases and desaturases (Monroig et al., 2013; Monroig and Kabeya, 2018; Monroig et al., 2022). More specifically, elongation of very long-chain fatty acid (Elovl) proteins are usually regarded as the rate-limiting enzyme in the FA elongation pathway (Jakobsson et al., 2006; Leonard et al., 2004). In vertebrates, eight different Elovl-like proteins (Elovl1-8) have been identified, with Elovl2, Elovl4 and Elovl5 playing well-established roles in LC-PUFA biosynthesis (Jakobsson et al., 2006; Leonard et al., 2004; Monroig et al., 2022). In invertebrates, current evidence confirms that several members of the Elovl family exist but the phylogenetic relationship with vertebrate homologues remains unclear in most instances (Monroig et al., 2022). Clearly though, both Elovl4 and Elovl2/5, an ancient form of the vertebrate’s Elovl2 and Elovl5, have been reported in molluscs (Monroig et al., 2016a, Monroig et al., 2016b). Within crustaceans, Elovl4 has been cloned and functionally characterised from the crabs Scylla olivacea (Ting et al., 2020) and Portunus trituberculatus (Sun et al., 2020), the gammarid Echinogammarus marinus (Ribes-Navarro et al., 2021) and the harpacticoid copepod Tigriopus californicus (Kabeya et al., 2021). Interestingly, some of the above studies report on further elongase enzymes and, along Elovl4, S. olivacea has two further Elovl-like sequences, E. marinus has two and T. californicus has five (Mah et al., 2019; Ting et al., 2020; Sun et al., 2020; Ribes-Navarro et al., 2021; Kabeya et al., 2021). With regard to fatty acyl desaturases, three major families have been shown to participate in animal LC-PUFA biosynthesis (Monroig and Kabeya, 2018; Monroig et al., 2022). First, the so-called methyl-end ( ω x) desaturases are enzymes that inserted a new double bond (unsaturation) between a pre-existing one and the methyl (–CH 3 ) terminus of the fatty acyl chain. Second, the “front-end desaturases” introduce a double bond between an existing one and the carboxylic group (–COOH) (Monroig et al., 2022). While both methyland front-end desaturases have been demonstrated to play key roles in the LC-PUFA biosynthesis in animals (Monroig and Kabeya, 2018; Monroig et al., 2022), a third family of fatty acyl desaturases termed as “first desaturases” are responsible to introduce the first double bond into saturated FA, typically at position 9 from –COOH, and hence termed as “Δ9 desaturases” (Hashimoto et al., 2008). While Δ9 desaturases do not typically participate in the biosynthetic pathways of LC-PUFA, investigations on arthropods including the house cricket Acheta domesticus (Zhou et al., 2008), the soldier beetle Chauliognathus lugubris (Haritos et al., 2012), the red flour beetle Tribolium castaneum (Haritos et al., 2014; Zhou et al., 2008) and the parasitoid wasp Nasonia vitripenis (Semmelmann et al., 2019), have shown that specific copies of the Δ9 desaturases co-existing in their genomes have evolved by acquiring some of the enzymatic capacities of methyl-end desaturases, namely Δ12 desaturation. Occurrence of neofunctionalised first desaturases in crustaceans remains to be confirmed but species with multiple copies of these genes are particularly interesting candidates. An early study by Schauer and Simpson (1985) reported that Artemia nauplii fed rice bran diet with 14 C labelled linoleic acid (LA) ([114 C] 18:2n-6) and α -linolenic acid (ALA) ([114 C]18:3n-3) showed the ability to bioconvert these compounds into LC-PUFA such as EPA and DHA. While bioconversion of ALA to longer-chain n-3 products like EPA and DHA had been reported in other crustaceans (e.g., Morris and Sargent, 1973; Henderson et al., 1981), biosynthesis of EPA and/or DHA from the n-6 precursor LA was unexpected (Schauer and Simpson, 1985). The methodological approach used by Schauer and Simpson (1985) could not completely rule out the possibility that those conversions were driven by microorganisms. Subsequently, Ito and Simpson (1996) tested the capacity of Artemia to bioconvert LA into several n-3 products under axenic conditions achieved by supplementing the culture with a mixture of antibiotics. The results confirmed the ability of Artemia to bioconvert LA into n-3 PUFA including ALA and EPA, suggesting the existences of desaturases enabling the interconversion between the n-6 and n-3 pathways, as well as front-end desaturases and PUFA Elovl. Yet specific genes responsible for such conversions remained to be elucidated. The present study aimed to perform an extensive search of potential genes involved in LC-PUFA biosynthesis in Artemia franciscana, arguably the most commonly used Artemia species in aquaculture worldwide. Using the bioinformatics resources available for A. franciscana (Jo et al., 2021), we performed a systematic search of genes encoding Elovl, as well as first, methyl-end and front-end desaturases, from A. franciscana, and performed the molecular and functional characterisation of the encoded enzymes. 2. Materials and methods 2.1. Retrieval of elongase and desaturase sequences from A. franciscana Putative elongase and desaturase transcripts were initially retrieved through tblastn searches against the National Center of Biotechnology Information (NCBI) (www.ncbi.nlm.nih.gov) Sequence Read Archive (SRA) contained in the A. franciscana BioProject (Accession No. PRJNA524488). The Daphnia pulex sequences EFX67847, EFX76305, EFX74345, EFX74342 and EFX88813 were used as queries to retrieve elongases from A. franciscana, whereas the D. pulex EFX60807, EFX70200, EFX82757 and EFX85062 sequences were used to retrieve A. franciscana desaturases. The retrieved raw reads were assembled using CLC Main Workbench 21 (Qiagen, Redwood City, CA, USA) to construct contigs containing putative full-length open reading frame (ORF) of the target genes. Subsequent release of the A. franciscana genome along with predicted transcriptome and protein database at the Antarctic Functional Genomics Program in Korea Polar Research Institute (https://antagen.kopri.re.kr/project/genome_info_iframe.php? Code=AF01, Accessed 12/07/2021) enabled to confirm the SRA M. Ramos-Llorens et al. Aquaculture 563 (2023) 738869 3 searches and complete full-length ORF sequences where necessary (Jo et al., 2021). In addition, protein databases were also searched to find amino acid (aa) sequences containing the Pfam domain “ELO” (PF01151) for elongases and “FA_desaturase” (PF00487) for desaturases, respectively, by using HMMER 3.1b1 with significance cut-off =0.01 (Finn et al., 2011; Mistry et al., 2021). Nine sequences containing ELO domain were retrieved, although one was excluded from functional analysis since it lacked the well-conserved histidine box (H-box) (HXXHH) (Hashimoto et al., 2008). Sequences were further confirmed by searching against Artemia SRA as above and completed when necessary. Consequently, a total of eight Elovl-like sequences (termed herein as “Elo1-8”) were found in A. franciscana. Regarding FA_desaturase domain, seven sequences were first identified. One of the sequences was confirmed as L-glutamate gammasemialdehyde dehydrogenase by blastp search against nr database in NCBI and hence disregarded. Three other sequences were also excluded from further analysis since, despite having three H-box, their aa composition did not fit the criteria of first, methyl-end or front-end desaturases according to Hashimoto et al. (2008). Finally, three sequences from A. franciscana thereafter referred to as “Des1-3” met the criteria set by Hashimoto et al. (2008) for desaturases and were subsequently considered for further analysis. 2.2. Sample collection, total RNA extraction and cDNA synthesis A. franciscana cysts from the IATS-CSIC Artemia Cyst Bank (Amat et al., 2020) were hydrated in ddH 2 O and subsequently decapsulated in a bleach solution. A subsample (~100 mg WW) of hydrated and decapsulated cyst was used for extraction of total RNA using the Maxwell® 16LEV simplyRNA Purification kit according to the manufacturer’s instructions (Promega, Madison, WI, USA). Finally, 2 μ g of total RNA were reverse transcribed into complementary DNA (cDNA) using the Moloney Murine Leukemia Virus Reverse Transcriptase (MMLV RT) (Promega) following the manufacturer’s instructions. 2.3. Molecular cloning of the A. franciscana elongase and desaturase genes The ORF of the retrieved A. franciscana elongases (Elo1-8) and desaturases (Des1-3) were amplified by PCR using the high-fidelity Phusion® DNA polymerase (Thermo Fisher Scientific, Waltham, MA, USA) with primers containing restriction enzyme sites for further cloning into the yeast expression plasmid pYES2 (Thermo Fisher Scientific) (Table 1). The PCR conditions were as follows: initial denaturing at 98◦C for 3 min, 35 cycles of denaturation at 98◦C for 30 s, annealing at 55◦C for 30 s, extension at 72◦C for 45 s, and a final extension at 72◦C for 10 min. PCR products were run on a 1% agarose gel and purified using the Wizard® SV gel and PCR clean-up System kit (Promega). Next, the purified PCR products containing the ORF of A. franciscana elongases and desaturases were digested with the corresponding restriction enzymes according to manufacturer’s instructions (New England Biolabs, Ipswich, MA, USA) (Table 1). After purification (Wizard® SV gel and PCR clean-up System, Promega), the restricted PCR products were ligated (T4 DNA ligase, Promega) into the similarly restricted yeast expression vector pYES2 to generate the following constructs: pYES2-Elo1, pYES2Elo2, pYES2-Elo3, pYES2-Elo4, pYES2-Elo5, pYES2-Elo6 and pYES2Elo7 (Elo8 was discarded from the functional assays for the reasons described below) for elongases, and pYES2-Des1, pYES2-Des2 and pYES2-Des3 for desaturases. E. coli Top10’ competent cells transformed with each of the A. franciscana elongase and desaturase pYES2 constructs were grown in LB agar plates with ampicillin for selection of positive colonies. Subsequently, colonies were screened by PCR and growth in LB broth for plasmid preparation (GenElute™ Plasmid Miniprep Kit, Sigma-Aldrich). Correctness of the ORF sequences was confirmed by Sanger DNA sequencing (DNA Sequencing Unit, IBMCPUPV, Valencia, Spain) prior use in the functional characterisation assays in yeast. 2.4. Sequence analysis and phylogenetics of the A. franciscana elongases and desaturases The deduced aa sequences of the putative A. franciscana elongases and desaturases were, respectively, performed with Geneious Prime Software (v: 2022.1.1) and sequences aligned with Clustal Omega (default settings) as implemented in the software (Sievers et al., 2011; Kearse et al., 2012). Characteristic features of fatty acyl elongases and desaturases were identified following Hashimoto et al. (2008). Phylogenetic trees were constructed to compare the deduced aa sequences of the A. franciscana elongases and desaturases with homologues retrieved from a variety of animals using the maximum-likelihood (ML) method (Whelan et al., 2001). For elongases, representatives for all Elovl subfamilies (Elovl1-8) known to exist in vertebrates were selected for analysis (Xie et al., 2021; Monroig et al., 2022). Furthermore, elongase sequences from invertebrates with existing genome projects (e.g., Daphnia magna, Hyalella azteca) or aquaculture relevant species including crustaceans (e.g., S. olivacea, E. marinus) and molluscs (e.g., Octopus vulgaris) with functionally characterised elongases (e.g., Monroig et al., 2012; Horn et al., 2017; Poynton et al., 2018; Lee et al., 2019; Mah et al., 2019; Sun et al., 2020; Ting et al., 2020; Kabeya et al., 2021; Ribes-Navarro et al., 2021) were also included in the phylogenetic analysis. Similarly, desaturase sequences from different subtypes including methyl-end desaturases, front-end desaturases, “first” desaturases from a variety of animal species were selected for phylogenetic analysis. Firstly, aa sequences were aligned with MUSCLE algorithm (default settings) (Edgar, 2004) as indicated above. The phylogenetic tree comparing aa sequences was constructed using the ML method (Jones et al., 1992) using MEGA X (Kumar et al., 2018). Confidence in the resulting phylogenetic tree branch topology was measured by bootstrapping through 1000 iterations (Felsenstein, 1985). 2.5. Functional characterisation of the A. franciscana elongases and desaturases by heterologous expression in yeast Plasmid constructs containing the ORF of the A. franciscana elongases (pYES2-Elo1, pYES2-Elo2, pYES2-Elo3, pYES2-Elo4, pYES2-Elo5, pYES2-Elo6 and pYES2-Elo7) and desaturases (pYES2-Des1, pYES2Des2 and pYES2-Des3) were transformed into competent Saccharomyces cerevisiae INvSc1 cells (Thermo Fisher Scientific) using the S.c. EasyComp® Transformation kit (Thermo Fisher Scientific). Transformed yeast were cultured on S. c. minimal medium minus uracil (SCMM −ura ) agar plates at 30◦C for 3 d and, subsequently, one single colony from Table 1 Primers and their nucleotide sequences used in the present study. Restriction sites used for cloning into the yeast expression vector pYES2 are underlined. Primer name Sequence 5′–3′ Elo1F-HindIII CCCAAGCTTAATATGTCTCTGCTACAATACTGTA Elo1R-XbaI CCGTCTAGACTATTTCCTCTTTTTACTTCGAATTAC Elo2F-SacI CCCGAGCTCAAAATGGGTTTAATAAGTAACATGAT Elo2R-XbaI CCGTCTAGATTATTCCATTTTATGTAATCCAACAG Elo34F-HindIII CCCAAGCTTAGGATGTTTATGAGCTCACCA Elo34R-XbaI CCGTCTAGATCATTCAATTTTCGAATGCAATG Elo5F-SacI CCCGAGCTCAAAATGGATTACACGTTGGAC Elo5R-XbaI CCGTCTAGATTACTTCAACTTATTCGCTTTAGC Elo6F-SacI CCCGAGCTCAAGATGGATAACTTGACTGATGA Elo6R-XbaI CCGTCTAGACTATTCTTGTTTTACATTACATTGTGT Elo7F-BamHI CGGGATCCCGCAGCATGGATTTTCT Elo7R-XbaI GCTCTAGAGCCAATTTGCGGCTTAATCAACTT Des1F-EcoRI CCCGAATTCAGAATGAAAGGCAAAGAGCAATTATCA Des1R-XhoI CCGCTCGAGTTATTTTACTAGCACCTCAGAGCC Des2F-HindIII CCCAAGCTTACAATGTGCCAAAGAGAA Des2R-XbaI CCGTCTAGATCACCAATTAACTGG Des3F-HindIII CCCAAGCTTAAGATGTATTCCACTATA Des3R-XbaI CCGTCTAGATTATTTCACCCGTGC M. Ramos-Llorens et al. Aquaculture 563 (2023) 738869 4 each plate was grown in SCMM −ura broth at 30◦C for 2 d until reaching an OD 600 between 8 and 10. Next, subcultures of 5 ml with a starting cell density of OD 600 =0.4 were set in each of the 150 ml Erlenmeyer flasks used to study the specific elongase or desaturase functions. After 4 h of constant shacking at 30◦C, transgenic yeast cultures were supplemented with 2% (w/v) galactose to induce gene expression, as well as one single polyunsaturated fatty acid (PUFA) substrate. Based on the LC-PUFA biosynthetic pathways of invertebrates (Monroig and Kabeya, 2018), potential elongase substrates, namely ALA (18:3n-3), LA (18:2n-6), stearidonic acid (18:4n-3), γ-linolenic acid (18:3n-6), eicosapentaenoic acid (EPA, 20:5n-3), arachidonic acid (ARA, 20:4n-6), docosapentanoic acid (22:5n-3) and docosatetranoic acid (22:4n-6), were supplemented to yeast expressing the A. franciscana elongases. Final concentrations of the exogenously added PUFA substrates were 0.5 mM (C 18 ), 0.75 mM (C 20 ) and 1.0 mM (C 22 ), as uptake efficiency decreases with increasing chain length. Since it is well established that Δ9-like desaturases do not typically recognise PUFA as substrates (Castro et al., 2016), the functional characterisation of the A. franciscana desaturases was assayed following a different approach to that described above for elongases. First, triplicated subcultures of control yeast transformed with the empty pYES2 vector and therefore only expressing the S. cerevisiae endogenous OLE1 (Δ9 desaturase) were compared with transgenic yeast that, along the S. cerevisiae endogenous OLE1, expressed one of the A. franciscana desaturases (Des1, Des2 and Des3). Furthermore, potential activities as Δ15 desaturase were assessed by growing transgenic yeast expressing Des1, Des2 and Des3, in the presence of LA (18:2n-6) supplemented at 0.5 mM. For both elongase and desaturase assays, yeast subcultures were maintained at 30◦C and constant shaking for 48 h after galactose induction. Yeast contained in each of the 5 ml subcultures were harvested by centrifugation at 1500g, washed twice in ddH 2 0, homogenised in chloroform/methanol (2:1, v/v) containing 0.01% (w/v) butylated hydroxytoluene (BHT) as antioxidant, and stored at −20◦C for further analysis. Yeast culture reagents including nitrogen base, uracil dropout medium, raffinose, Tergitol® and galactose, as well as BHT, were purchased from Sigma-Aldrich (St. Louis, MS, USA). Exogenously added FA substrates used for functional characterisation in yeast were purchased from Nu-Chek Prep Inc. (Elysian, Minnesota, USA) except 20:4n-6, which was from Cayman Chemical (Ann Arbor, Michigan, USA). 2.6. Fatty acid analysis from yeast Total lipids were extracted from yeast samples following the methodology by Folch et al. (1957) with modifications as described by RibesNavarro et al. (2021). Subsequently, total lipids were used to prepare fatty acid methyl esters (FAME) though acid catalysed transmethylation using the method of Christie (2003), and purified using thin layer chromatography on 20 ×20 cm plates (Merk KGaA, Darmstadt, Germany). FAME from the desaturases experiments were injected onto a Thermo gas chromatograph (Thermo Trace GC Ultra, Thermo Electron Corporation, Waltham, MA, USA) fitted with an on-column injection system, a FID detector, and a silica capillary column (30 m ×0.25 mm × 0.25 μ m film thickness, TR-WAX, Teknokroma, Spain), using helium as a carrier gas. FAME from the elongase assays were analysed using an Agilent 6850 Gas Chromatograph system coupled to a 5975 series MSD (Agilent Technologies, Santa Clara, CA, USA) equipped with a Sapiens-5MS (30 m ×0.25 μ m ×0.25 μ m) capillary column (Teknokroma). The proportion of the each PUFA substrate converted to the corresponding elongated products was calculated as [product areas/(- product areas +substrate area)] x 100. The activity of the A. franciscana desaturases was estimated by comparing the FA profiles (expressed as % of total FA) of control yeast (transformed with the empty pYES2) with those from yeast expressing the corresponding desaturase gene (i.e., transformed with pYES2-Des1, pYES2-Des2 or pYES2-Des3). The affinity of the A. franciscana desaturase for the exogenously added substrate LA was also tested (potential Δ15 desaturase activity). All solvents were analytical grade (>99%) and were purchased from Merk. 2.7. Statistics For the functional characterisation experiments of the A. franciscana Des1, Des2 and Des3, FA analyses from yeast samples were expressed as mean values ±standard deviation (n =3). Comparison of FA profiles from controls and yeast expressing the A. franciscana desaturases Des1, Des2 and Des3 was analysed with three indivitual Student’s t-tests. Comparisons of the means with p values less or equal than 0.05 were considered significantly different. All the statistical analyses were carried out using the SPSS statistical package (SPSS Inc., Chicago, IL, USA). 3. Results 3.1. Sequences and phylogeny of the A. franciscana elongases A total of eight Elovl-like sequences (Elo1-8) were retrieved from A. franciscana genomics databases and deposited into the NCBI GenBank with the accession numbers MZ262329 (Elo1), MZ262324 (Elo2), MZ262325 (Elo3), MZ262326 (Elo4), MZ262327 (Elo5), MZ262328 (Elo6), OM677865 (Elo7) and ON416870 (Elo8). The ORF of the herein termed Elo1 from A. franciscana consists of 993 base pairs (bp), encoding a putative protein of 330 aa. Moreover, the A. franciscana Elo2, Elo5, Elo6 and Elo7 ORF sequences consist of 990, 897, 909 and 807 bp, respectively, encoding putative proteins of 329, 298, 302 and 268 aa. Both Elo3 and Elo4 have an ORF of 1035 bp, encoding a putative protein of 344 aa. Finally, Elo8 consists of 864 bp encoding 287 aa. The analysis of the deduced aa sequences revealed that all eight Elovl-like sequences from A. franciscana contained the distinctive H-box (HXXHH) (Fig. 1). Moreover, the aa residues located in the N-terminal side of the H-box, used to distinguish PUFA and non-PUFA elongases (Hashimoto et al., 2008), showed some differences among the six A. franciscana elongases. The A. franciscana Elo1, Elo2, Elo5, Elo6 and Elo7 contained a glutamine (Q) at position −5 preceding the H-box, whereas Elo3 and Elo4 contained a histidine (H) in −5 position (Fig. 1). While both alternative residues (Q and H) satisfy the criteria established by Hashimoto et al. (2008) for PUFA elongases, the A. franciscana Elo8 contained a proline (P) at position −5 preceding the H-box (Fig. 1) and was therefore excluded from further functional characterisation assay. The aa sequences of the A. franciscana Elovl-like were compared to elongases from a variety of animals by constructing a phylogenetic tree (Fig. 2). Results showed that the A. franciscana Elo1 protein grouped together with elovl4 orthologues from vertebrates and invertebrates including crustaceans like H. azteca and D. magna (Fig. 2). These results confirm that the newly cloned A. franciscana Elo1 is an orthologue of Elovl4. Another well supported cluster containing the A. franciscana Elo2, Elo3, Elo4, Elo5 and Elo6, along with elongases from crustaceans such as S. olivacea, E. marinus, T. californicus, etc., was also identified in the tree (Fig. 2). This branch contained sequences belonging to Elovl1 and Elovl7 from a range of vertebrate species (Fig. 2). The A. franciscana Elo7 grouped together with Elovl8 sequences characterised from teleosts (e.g., Danio rerio) (Fig. 2). Finally, the A. franciscana Elo8 clustered together with Elovl6-like sequences. None of the eight A. franciscana elongases grouped with sequences forming a branch containing the PUFA elongases Elovl2 and Elovl5 from vertebrates, and the molluscan Elovl2/5 (Fig. 2). 3.2. Sequence and phylogenetics of A. franciscana desaturases The ORF of the three A. franciscana desaturases termed as Des1, Des2 and Des3 contain 1005, 1029 and 1053 bp, respectively, encoding putative proteins of 334, 342 and 350 aa (Fig. 3). The newly cloned desaturase sequences from A. franciscana were deposited into the NCBI GenBank with the following accession numbers: MZ254648 (Des1), MZ254649 (Des2) and MZ254650 (Des3). The three sequences M. Ramos-Llorens et al. Aquaculture 563 (2023) 738869 5 contained three distinct H-box characteristic of desaturases (Hashimoto et al., 2008). Specifically, the A. franciscana desaturases H-box were HXXXXH (H-box1), HXXHH (H-box2) and HXXHH (H-box3) (Fig. 3). Composition of residues within H-box1 (HRLWXH), H-box2 (HRVHH) and H-box3 (HNXHH) are consistent with “First” desaturases (Hashimoto et al., 2008). Consistently, our phylogenetic analysis showed that the A. franciscana Des1-3 proteins grouped together with first desaturases mostly represented by stearoyl-CoA desaturase orthologues from a variety of animals, and distantly from other desaturase subfamilies including methyl-end and front-end desaturases (Fig. 4). 3.3. Roles of A. franciscana elongases in LC-PUFA biosynthesis The ability of the A. franciscana Elovl-like enzymes for PUFA elongation was examined by incubating transgenic yeast expressing their ORF in the presence of PUFA substrates. Our results showed that the elongation ability towards exogenously supplemented PUFA varied among the A. franciscana elongases (Table 2). First, both Elo2 and Elo5 were able to convert all PUFA assayed (18:3n-3, 18:2n-6, 18:4n-3, 18:3n-6, 20:5n-3, 20:4n-6, 22:5n-3 and 22:4n-6) to the corresponding 2‑carbon elongated products (Table 2). Moreover, the A. franciscana Elo1 and Elo6 had in common the ability to elongate all the C 18 PUFA substrates assayed as well as 20:5n-3, with 22:5n-3 being elongated to 24:5n-3 by Elo1 but not Elo6 (Table 2). The A. franciscana Elo3 and Elo4 recognised only two C 18 PUFA as substrates, namely 18:4n-3 and 18:3n6 for Elo3, and 18:2n-6 and 18:3n-6 for Elo4, with no activity detected towards any C 20 and C 22 PUFA (Table 2). Finally, the A. franciscana Elo7 showed ability to elongate all C 18 and C 20 PUFA substrates up to C 22 PUFA products, being generally more active over n-3 PUFAs than n-6 PUFAs. Additionally, some elongases such as Elo1 and Elo7 mediated stepwise elongation reactions beyond the first elongation product (Table 2). Thus, Elo1 elongated the exogenously supplied 18:3n-3 to 20:3n-3, the latter being further elongated to 22:3n-3, whereas supplementation of 18:4n-3 and 18:3n-6 to yeast expressing the A. franciscana Elo7 led to the synthesis of the corresponding C 20 and C 22 elongated products. 3.4. Functions of A. franciscana desaturases The desaturation activity of the A. franciscana desaturases (Des1-3) was investigated by growing transgenic yeast expressing their ORF in the absence of exogenously supplemented FA substrates and in the presence of exogenously supplied LA. Comparison of the FA profiles of control yeast and transgenic yeast expressing each of the A. franciscana desaturases and grown in the absence of any supplemented added FA substrate, showed that yeast expressing the A. franciscana Des1 and Des2 had a significantly higher content of the Δ9 desaturation product 18:1n9 (OA) (Table 3). Although only significant for Des1, reduced levels of the Δ9 desaturation substrate 18:0 were detected in yeast expressing Des1 and Des2 compared to control yeast (Table 3). Both Des2 and Des3 showed a significant increase of 18:2n-6 (LA) as compared to the control yeast, suggesting these genes have Δ12 desaturase activity (Table 3). Finally, no additional peaks corresponding to desaturation products were detected in transgenic yeast grown in the presence of LA, suggesting that any of the retrieved desaturase from A. franciscana has Δ15 desaturase activity (data not shown). 4. Discussion Arguably one of the most striking results obtained in the present study is the remarkably high number of PUFA elongases in A. franciscana. With the exception of Elo8 (phylogenetically identified as an orthologue of Elovl6), A. franciscana possess seven Elovl-like sequences whose deduced protein sequences satisfy the criteria of PUFA elongases according to the classification proposed by Hashimoto et al. (2008). Our phylogenetic analyses show that the A. franciscana Elo1 protein grouped together with elovl4-like sequences from a variety of animal species, confirming that this gene is an orthologue of elovl4. Elovl4-like enzymes have been found in molluscs (Monroig et al., 2017; Ran et al., 2019), as well as other crustaceans (e.g., Ribes-Navarro et al., Fig. 1. Alignment of the deduced amino acid sequences of the A. franciscana elovl-like cDNAs investigated herein (Elo1, Elo2, Elo3, Elo4, Elo5, Elo6, Elo7 and Elo8), performed with Geneious Prime Software (v: 2022.1.1) and sequences aligned with Clustal Omega (default setting) as implemented in the software (Sievers et al., 2011; Kearse et al., 2012). Elongase conserved motif histidine box (HXXHH) (Hashimoto et al., 2008) is underlined and −5 histidine box position is marked by an arrow. M. Ramos-Llorens et al. Aquaculture 563 (2023) 738869 6 2021). Also consistent with current evidence available for crustaceans (Monroig et al., 2022), none of the elovl-like sequences found in A. franciscana is phylogenetically related to the invertebrate PUFA elongase Elovl2/5 characterised in molluscs (Liu et al., 2013; Monroig et al., 2012, 2016a; Zhang et al., 2018) and amphioxus (Monroig et al., 2016b). However, such an apparent loss of PUFA elongase capacity associated to the absence of Elovl2/5 is clearly compensated for in Artemia by the occurrence of the particularly high number of PUFA Elovl-like sequences alluded to above. Among them, the cluster containing the A. franciscana Elo2, Elo3, Elo4, Elo5 and Elo6 also includes previously described elongases from the crustaceans S. olivacea (Mah et al., 2019), E. marinus (Ribes-Navarro et al., 2021) and T. californicus (Kabeya et al., 2021), as well as newly in silico identified but yet uncharacterised elongases from H. azteca, D. magna, and T. newberryi. The resolution of the tree does not allow us to unequivocally clarify the phylogenetics of the A. franciscana Elo2-6 elongases but, more clearly though, they appear to be related to Elovl1 and Elovl7 according to the presence of vertebrate Elovl1 and Elovl7 sequences in the same cluster. Interestingly, whereas one sole representative of this cluster can be found in other crustaceans such as the gammarid E. marinus (Ribes-- Navarro et al., 2021) and two representatives can be found in the genome of H. azteca (Poynton et al., 2018), A. franciscana has five Fig. 2. Phylogenetic tree comparing A. franciscana Elo1, Elo2, Elo3, Elo4, Elo5, Elo6, Elo7 and Elo8 with elongase proteins from other organisms. The tree was constructed using the maximum likelihood method (Jones et al., 1992) using MEGA-X software (Kumar et al., 2018). The numbers represent the frequencies (%) with which the tree topology presented was replicated after 1000 iterations (Felsenstein, 1985). M. Ramos-Llorens et al. Aquaculture 563 (2023) 738869 7 elongases belonging to the same cluster, suggesting an enhanced elongation capacity compared to other crustaceans. Furthermore, the newly cloned A. franciscana Elo7 is phylogenetically related to Elovl8, the most recently described member of the Elovl family in vertebrates and found to play important roles in PUFA elongation in fish (Li et al., 2020; Sun et al., 2021). The categorisation of the A. franciscana Elo1-7 as PUFA elongases based on specific sequence features according to Hashimoto et al. (2008), was confirmed by our functional assays in yeast. Seven distinct elongases have shown to be active towards a variety of PUFA substrates ranging from C 18 to C 22 and thus enabling A. franciscana to perform all elongation reactions from C 18 PUFA precursors up to the corresponding C 24 products. The conversions exhibited by several A. franciscana Elovl when expressed in yeast were generally low as previously reported for other crustacean elongases characterised in the same yeast heterologous system (e.g., Ribes-Navarro et al., 2021). Whereas the yeast system used herein cannot establish the activity that the characterised enzymes have in vivo, judging from the conversions detected in our functional assays is tempting to speculate that the A. franciscana Elo7, an orthologue of Elovl8, is a major elongase involved in PUFA elongation in A. franciscana. Like the Elovl8 from the rabbitfish Siganus canaliculatus, the A. franciscana Elovl8 showed elongation capacity towards C 18 and C 20 PUFA substrates (Li et al., 2020). Contributions by Elovl8 and other Elovl enzymes present in A. franciscana are likely accounting for the elongation capacity reported elsewhere in the literature for Artemia sp. determined by comparison of body lipid composition after feeding experiments (Zhukova et al., 1998) or direct biochemical evidence (Schauer and Simpson, 1985; Ito and Simpson, 1996). However, the latter studies also suggested that Artemia has desaturation capacity that, in combination with elongases, enable the synthesis of LC-PUFA such as EPA. Our extensive characterisation of the fatty acyl desaturase repertoire from A. franciscana does not support those findings. Retrieval of desaturases with potential roles in the A. franciscana LCPUFA biosynthesis resulted in three sequences that fulfilled the criteria by either first, methyl-end or front-end desaturases established by Hashimoto et al. (2008). All three sequences were phylogenetically related to first desaturases such as stearoyl-CoA desaturases that typically have Δ9 desaturase activity towards saturated FA such stearic acid (18:0) to produce monounsaturated FA like OA (18:1n-9) (Monroig et al., 2022). Consistently, the herein characterised A. franciscana Des1 and Des2 enzymes were Δ9 desaturases catalysing the biosynthesis of OA in A. franciscana. Interestingly, the A. franciscana Des3 did not have an apparent Δ9 desaturase activity but rather Δ12 activity since a small but significant increase of LA (18:2n-6) was detected in yeast expressing Des3 compared to control yeast. These results suggest that, in the absence of methyl-end desaturases denoted from our searches, one of the A. franciscana first desaturases has undergone a neofunctionalisation from Δ9 desaturase activity to Δ12. The A. franciscana Des3 therefore represents, to the best of our knowledge, the first case of a Δ9-like enzyme acquiring Δ12 desaturase activity in crustaceans, an evolutionary adaptation reported in other terrestrial arthropods like N. vitripenis in which specific copies of their Δ9 desaturases have acquired Δ12 desaturation ability to support the production of LA-derived sex pheromones in males (Semmelmann et al., 2019). While the herein reported Δ12 desaturase capacity enables the introduction of a second unsaturation and hence the de novo biosynthesis of PUFA in A. franciscana (Monroig et al., 2022), the abovementioned absence of genome-anchored methyl-end desaturases in this species precludes the biosynthesis of ALA from LA (Δ15 desaturation). Consequently, our results do not support the previous findings by Schauer and Simpson (1985) as well as Ito and Simpson (1996), who observed the ability of Artemia for bioconverting LA into ALA (18:3n-3). Rather, the results collected in the present study are in agreement with those reported more recently by Reis et al. (2017) in which A. franciscana metanauplii did not show desaturation products when incubated with radiolabelled LA. Importantly, the experiments of Ito and Simpson (1996) with radiolabelled LA resulted, along ALA, in the production of other metabolic products including 18:4n-3 and EPA (20:5n-3). While the abilities demonstrated for A. franciscana Elovl enzymes might account for the production of EPA from C 18 PUFA precursors, no evidence supporting the presence of front-end desaturases with Δ6 and Δ5 activities explaining the presence of 18:4n-3 and EPA has been found in our study. Collectively, our results strongly suggest that, unlike PUFA elongation, A. franciscana has little desaturase capacity and is therefore unable to carry out most of the desaturase reactions required for the biosynthesis of LC-PUFA. While symbionts have been suggested to partly contribute to LC-PUFA found in A. franciscana (Wache and Laufer, 1998), the presence of LC-PUFA in A. franciscana lipids appears to be mostly driven by diet available at biotopes inhabited by this crustacean. In agreement, experiments carried out at mesocosm scale revealed that the FA profile of Artemia is mainly phenotypically driven (Ruiz et al., 2007), with diet playing a preponderant role in the final FA profile of both biomasses and cysts (Ruiz et al., 2008). Indeed, the abovementioned freshwater-marine-type classification of the different batches of cysts and nauplii for aquaculture purposes reflects a diatomrich impact of diet in the fatty acid profile of marine-type Artemia with the high 16:0/16:1 ratio, low 18:3n-3 and high 20:5n-3 acting as biomarkers (Ruiz et al., 2008). Thus the phenotypic impact of diet is consistent with the results of the present work indicating the lack of LCPUFA biosynthetic capacity of the genus based on its gene repertoire. In conclusion, we have identified eight elongases and three fatty acyl desaturases from A. franciscana. The deduced sequences of seven out of eight elongases found in A. franciscana have characteristics of PUFA elongase enzymes. Functional analyses in yeast confirmed A. franciscana can perform all the elongation reactions involved in the LC-PUFA biosynthetic pathways, with Elovl8 (termed herein as “Elo7”) playing an apparent prominent role. However, the repertoire of fatty acyl desaturases present in A. franciscana is limited to three genes of the socalled “first desaturases”, two of which (Des1 and Des2) have the expected Δ9 desaturase activity enabling the biosynthesis of 18:1n-9, whereas Des3 showed Δ12 activity enabling the biosynthesis of 18:2n6. Neither methyl-end nor front-end desaturases were found in Fig. 3. Alignment of A. franciscana Des1, Des2 and Des3 protein sequences performed with Geneious Prime Software (v: 2022.1.1) and sequences aligned with Clustal Omega (default setting) as implemented in the software (Sievers et al., 2011; Kearse et al., 2012). Conserved histidine boxes for first desaturases ([HXXXXH], [HXXHH] and [HXXHH]) (Hashimoto et al., 2008) are underlined. M. Ramos-Llorens et al. Aquaculture 563 (2023) 738869 8 Fig. 4. Phylogenetic tree comparing A. franciscana Des1, Des2 and Des3 with desaturase sequences from a variety of animal species. Sequences from different enzyme types, namely first desaturases, methyl-end desaturases and front-end desaturases, were analysed. The tree was constructed using the maximum likelihood method (Jones et al., 1992) by MEGA-X software (Kumar et al., 2018). The numbers represent the frequencies (%) with which the tree topology presented was replicated after 1000 iterations (Felsenstein, 1985). M. Ramos-Llorens et al. Aquaculture 563 (2023) 738869 9 A. franciscana. Overall, the present study demonstrated that A. franciscana has limited capacity for LC-PUFA biosynthesis due to the lack of key desaturases in its genome, and confirmed that current enrichment procedures ensuring the exogeneous provision of essential fatty acids to marine larvae are mandatory to prevent deficiency symptoms. CRediT authorship contribution statement Marc Ramos-Llorens: Investigation, Methodology, Formal analysis, Writing – original draft, Visualization. Alberto Ribes-Navarro: Investigation, Methodology, Formal analysis, Writing – original draft, Visualization. Juan C. Navarro: Conceptualization, Investigation, Supervision, Writing – review & editing, Funding acquisition. Francisco Hontoria: Conceptualization, Investigation, Supervision, Writing – review & editing, Funding acquisition. Naoki Kabeya: Conceptualization, Investigation, Methodology, Writing – review & editing. ´ Oscar Monroig: Conceptualization, Investigation, Supervision, Writing – review & editing, Funding acquisition. Declaration of Competing Interest The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Oscar Monroig reports financial support was provided by Agencia Estatal de Investigaci´ on. Data availability Data will be made available on request. Acknowledgements This study was funded through the project IMPROMEGA Agencia Espa˜ nola de Investigaci´ on, Spain, grant no. RTI2018-095119-B-100, MCIU/AEI/ FEDER/UE / MCIN/AEI/10.13039/501100011033/ and FEDER “A way to make Europe”. References Amat, F., Hontoria, F., Var´ o, I., Navarro, J.C., 2020. Artemia (Crustacea, Branchiopoda, Anostraca) cyst bank (IATS, CSIC). V1.0. Instituto de Acuicultura Torre de la Sal (IATS, CSIC). https://doi.org/10.15470/s7zzqk. Dataset/Sampling event. Ando, Y., Kotake, M., Ota, T., 1997. Lipids and fatty acids in Artemia nauplii enriched with fish oil triacylglycerols containing docosahexaenoic acid in different positional distribution patterns. Fish. Sci. 63, 605–609. https://doi.org/10.2331/ fishsci.63.605. Calder, P.C., 2018. Very long-chain n-3 fatty acids and human health: fact, fiction and the future. Proc. Nutr. Soc. 77, 52–72. https://doi.org/10.1017/ S0029665117003950. Castro, L.F.C., Tocher, D.R., Monroig, ´ O., 2016. Long-chain polyunsaturated fatty acid biosynthesis in chordates: insights into the evolution of fads and elovl gene repertoire. Prog. Lipid Res. 62, 25–40. https://doi.org/10.1016/j. plipres.2016.01.001. Christie, W.W., 2003. Lipid Analysis, Isolation, Separation, Identification and Structural Analysis of Lipids, 3rd. The Oily Press., Bridgwater. Coutteau, P., Sorgeloos, P., 1997. Manipulation of dietary lipids, fatty acids and vitamins in zooplankton cultures. Freshw. Biol. 38, 501–512. https://doi.org/10.1046/ j.1365-2427.1997.00239.x. Edgar, R.C., 2004. MUSCLE: a multiple sequence alignment method with reduced time and space complexity. BMC Bioinforma. 5, 113. https://doi.org/10.1186/14712105-5-113. Felsenstein, J., 1985. Confidence limits on phylogenies with a molecular clock. Syst. Zool. 34, 152–161. https://doi.org/10.2307/sysbio/34.2.152. Finn, R.D., Clements, J., Eddy, S.R., 2011. HMMER web server: interactive sequence similarity searching. Nucleic Acids Res. 39, W29–W37. https://doi.org/10.1093/ nar/gkr367. Folch, J., Lees, M., Sloane Stanley, G.H., 1957. A simple method for the isolation and purification of total lipides from animal tissues. J. Biol. Chem. 226, 497–509. Guinot, D., Monroig, T., Hontoria, F., Amat, F., Var´ o, I., Navarro, J.C., 2013a. Enriched on-grown Artemia metanauplii actively metabolise highly unsaturated fatty acid-rich phospholipids. Aquaculture 412-413, 173–178. https://doi.org/10.1016/j. aquaculture.2013.07.030. Guinot, D., Monroig, O., Navarro, J.C., Var´ o, I., Amat, F., Hontoria, F., 2013b. Enrichment of Artemia metanauplii in phospholipids and essential fatty acids as a diet for common octopus (octopus vulgaris) paralarvae. Aquac. Nutr. 19, 837–844. https://doi.org/10.1111/anu.12048. Haritos, V.S., Horne, I., Damcevski, K., Glover, K., Gibb, N., Okada, S., Hamberg, M., 2012. The convergent evolution of defensive polyacetylenic fatty acid biosynthesis genes in soldier beetles. Nat. Commun. 3, 1150. https://doi.org/10.1038/ ncomms2147. Haritos, V.S., Horne, I., Damcevski, K., Glover, K., Gibb, N., 2014. Unexpected functional diversity in the fatty acid desaturases of the flour beetle Tribolium castaneum and identification of key residues determining activity. Insect Biochem. Mol. Biol. 51, 62–70. https://doi.org/10.1016/j.ibmb.2014.05.006. Hashimoto, K., Yoshizawa, A.C., Okuda, S., Kuma, K., Goto, S., Kanehisa, M., 2008. The repertoire of desaturases and elongases reveals fatty acid variations in 56 eukaryotic genomes. J. Lipid Res. 49, 183–191. https://doi.org/10.1194/jlr.M700377-JLR200. Henderson, R.J., Sargent, J.R., Falk-Petersen, S., 1981. Lipogenesis in the arctic euphausiid Thysanoessa inermis. Mar. Biol. 63, 235–240. https://doi.org/10.1007/ BF00395992. Horn, R.L., Ramaraj, T., Devitt, N.P., Schilkey, F.D., Cowley, D.E., 2017. De novo assembly of a tadpole shrimp (Triops newberryi) transcriptome and preliminary differential gene expression analysis. Mol. Ecol. Resour. 17, 161–171. https://doi. org/10.1111/1755-0998.12555. Ito, M.K., Simpson, K.L., 1996. The biosynthesis of ω 3 fatty acids from 18:2 ω 6 in Artemia spp. Comp. Biochem. Physiol. B Biochem. Mol. Biol. 115, 69–76. https://doi.org/ 10.1016/0305-0491(96)00091-0. Jakobsson, A., Westerberg, R., Jacobsson, A., 2006. Fatty acid elongases in mammals: their regulation and roles in metabolism. Prog. Lipid Res. 45, 237–249. https://doi. org/10.1016/j.plipres.2006.01.004. Janssen, C.I.F., Kiliaan, A.J., 2014. Long-chain polyunsaturated fatty acids (LCPUFA) from genesis to senescence: the influence of LCPUFA on neural development, aging, Table 2 Functional characterisation of A. franciscana elongases. Conversions of exogenously supplemented polyunsaturated fatty acid (FA) substrates were calculated according to the formula [product areas/ (product areas +substrate area)] x 100. FA substrate Product Elo1 Elo2 Elo3 Elo4 Elo5 Elo6 Elo7 Activity 18:3n-3 20:3n-3 0.75* 0.57 n.d. n.d. 0.45 0.45 8.08 C18 ➔ C20 18:2n-6 20:2n-6 0.29 0.39 n.d. 0.59 0.51 0.84 3.92 C18 ➔ C20 18:4n-3 20:4n-3 1.09 0.92 0.36 n.d. 0.58 2.47 45.50* C18 ➔ C20 18:3n-6 20:3n-6 0.33 0.52 1.66 0.30 0.43 1.55 18.47* C18 ➔ C20 20:5n-3 22:5n-3 0.12 0.12 n.d. n.d. 0.03 0.24 9.01 C20 ➔ C22 20:4n-6 22:4n-6 n.d. 0.11 n.d. n.d. 0.06 n.d. 2.34 C20 ➔ C22 22:5n-3 24:5n-3 0.12 0.66 n.d. n.d. 0.07 n.d. n.d. C22 ➔ C24 22:4n-6 24:4n-6 n.d. 0.25 n.d. n.d. 0.30 n.d. n.d. C22 ➔ C24 n.d., not detected. * Conversions of stepwise reactions from C 20 products are included. Table 3 Fatty acid (FA) composition of yeast from the functional characterisation assays of the A. franciscana desaturases Des1, Des2 and Des3. Control consisted of yeast S. cerevisiae transformed with empty pYES2 vector. Significant differences of yeast expressing the A. franciscana desaturases (treatments Des1, Des2 and Des3) against control yeast (t-test) are indicated according to the level of significance (* p <0.05, **p <0.01, ***p <0.001). FA Control Des1 Des2 Des3 16:0 23.1 ±0.5 20.3 ±1.1 ** 24.1 ±1.2 23.7 ±1.0 16:1n-9 0.5 ±0.0 0.3 ±0.0 * 0.5 ±0.0 0.5 ±0.0 16:1n-7 42.6 ±0.4 32.6 ±1.1 *** 38.4 ±1.5 ** 43.8 ±1.2 18:0 8.7 ±0.2 5.4 ±0.3 *** 8.4 ±0.9 9.5 ±0.6 ** 18:1n-9 23.8 ±0.3 40.01 ±0.7 *** 27.2 ±0.6 *** 21.2 ±0.4 *** 18:1n-7 1.4 ±0.1 1.3 ±0.1 * 1.2 ±0.1 ** 1.2 ±0.0 *** 18:2n-6 0.1 ±0.0 0.1 ±0.0 0.2 ±0.1 * 0.2 ±0.0 ** M. Ramos-Llorens et al.