Functional characterization and differential nutritional regulation of putative Elovl5 and Elovl4 elongases in large yellow croaker (Larimichthys crocea)

In the present study, two elongases, Elovl4 and Elovl5, were functionally characterized and their transcriptional regulation in response to n-3 LC-PUFA administration were investigated in vivo and in vitro. We previously described the molecular characterization of croaker elovl5. Here, we report the full-length cDNA sequence of croaker elovl4, which contained 1794 bp (excluding the polyA tail), including 909 bp of coding region that encoded a polypeptide of 302 amino acids possessing all the characteristic features of Elovl proteins. Functional studies showed that croaker Elovl5, displayed high elongation activity towards C18 and C20 PUFA, with only low activity towards C22 PUFA. In contrast, croaker Elovl4 could effectively convert both C20 and C22 PUFA to longer polyenoic products up to C34. n-3 LC-PUFA suppressed transcription of the two elongase genes, as well as srebp-1 and lxrα, major regulators of hepatic lipid metabolism. The results of dual-luciferase reporter assays and in vitro studies both indicated that the transcriptions of elovl5 and elovl4 elongases could be regulated by Lxrα. Moreover, Lxrα could mediate the transcription of elovl4 directly or indirectly through regulating the transcription of srebp-1. The above findings contribute further insight and understanding of the mechanisms regulating LC-PUFA biosynthesis in marine fish species.

Characteristically, the croaker elovl4 deduced protein contained one histidine box (HXXHH), five putative membrane-spanning domains, and single lysine and arginine residues (RXKXX) at the carboxyl terminus (Fig. 1). Phylogenetic tree analysis showed that the croaker Elovl4 clustered with several other Elovl4-like sequences from fish and mammals, and more distantly with Elovl2 and Elovl5 from these vertebrate lineages (Fig. 2). The results strongly suggested that the newly cloned cDNA encoded an Elovl4 elongase.
Functional characterization of the Elovl4 and Elovl5 elongases from L. crocea. The functions of croaker Elovl4 and Elovl5 were characterized by determining the FA profiles of S. cerevisiae transformed with pYES2-Elo4 or pYES2-Elo5, respectively, and grown in the presence of potential FA substrates. For Elovl4, its role in the biosynthesis of saturated VLC-FA was studied by comparing the saturated FA (>C 24 ) profiles of yeast transformed with pYES2-Elo4 and pYES2 (control) ( Table 1). No significant differences were found in the content of 24:0, 26:0, 28:0, 30:0 or 32:0, indicating that the croaker Elovl4 does not have a role in the biosynthesis of saturated VLC-FA. In order to investigate the role of the L. crocea Elovl4 in the biosynthesis of VLC-PUFA, pYES2-Elo4 transformed yeast were grown in the presence of C 18 (18:4n-3 and 18:3n-6), C 20 (20:5n-3 and 20:4n-6) and C 22 (22:5n-3, 22:4n-6 and 22:6n-3) substrates. The results confirmed that croaker Elovl4 was able to elongate all PUFA substrates producing in most cases polyenes of up to 32 or 34 carbons, except for DHA that was only elongated to 24:6n-3 (Table 2). Interestingly, the elongation towards C 28 and C 30 substrates showed the highest conversion values (%) regardless of the initial PUFA substrate. However, despite limited activity towards DHA, the croaker Elovl4 could elongate both 20:5n-3 and 22:5n-3 to 24:5n-3, the C 24 substrate for DHA synthesis.
Tissue distribution of elovl4. The distribution of elovl4 mRNA in L. crocea was detected in several tissues, including eye, brain, testis, heart, liver, kidney, spleen, stomach, intestine and muscle. Gene expression analysis revealed that the highest levels of elovl4 transcript were found in eye, follow by brain and testis (Fig. 3).
Nutritional regulation of the L. crocea elovl4 and elovl5 and transcription factors: in vivo and in vitro trials. The expression of elovl4 and elovl5 from fish fed the experimental diets containing varying levels of LC-PUFA is shown in Fig. 4A. The results showed that the expression of both genes was down-regulated in fish fed diets containing both high and moderate n-3 LC-PUFA levels in comparison to diets with low n-3 LC-PUFA Figure 1. Comparison of the deduced amino acid (AA) sequences of Elovl4 from large yellow croaker, other fish, mouse and human. The AA sequences were aligned using ClustalX, and identity/similarity shading was based on a 75% identity threshold. Identical residues are shaded black and similar residues are shaded grey. Indicated are the conserved HXXHH histidine box motif, five (I-V) putative membrane-spanning domains and the ER retrieval signal.
(P < 0.05). In vitro assays performed with hepatocyte primary cultures prepared from fish fed a standard diet indicated that the expression of elovl4, but not elovl5, was down-regulated by DHA supplementation (P < 0.05) (Fig. 4B). Additionally, supplementation of EPA to hepatocytes suppressed the expression of both elongases (P < 0.05) (Fig. 4C).
With regards to the regulation of transcription factors, dietary n-3 LC-PUFA significantly decreased the expression of srebp-1 and lxrα (P < 0.05) (Fig. 5A). The in vitro studies clearly showed that both DHA and EPA significantly suppressed the transcription of srebp-1 and lxrα (P < 0.05) ( Fig. 5B and C).
Transcriptional regulation of the L. crocea elovl4 and elovl5 by Lxrα and Srebp-1. Over-expression of Lxrα significantly increased the transcription of elovl4 (P < 0.05), but the expression of elovl4 was significantly suppressed through the inhibition of Srebp-1 (P < 0.05) in yellow croaker hepatocytes (Fig. 6A). These results indicated that both lxrα and srebp up-regulated elovl4. The addition of DHA or EPA significantly reduced the abovementioned up-regulation effects of lxrα and srebp (P < 0.05) (Fig. 6A). In addition, the inhibition of Srebp-1 significantly decreased the regulatory role of LXRα on the transcription of elovl4 (P < 0.05) (Fig. 6A).

Figure 2.
Phylogenetic tree comparing the large yellow croaker Elovl4 with elongase proteins from other organisms. The tree was constructed using the Neighbor Joining method [45] using MEGA4. The numbers represent the frequencies (%) with which the tree topology presented was replicated after 1000 iterations. Regarding the transcription of elovl5, over-expression of lxrα significantly increased the transcription of elovl5 (P < 0.05), while the inhibition of Srebp-1 had no significant effect on elovl5 expression (P > 0.05) (Fig. 6B). Moreover, the addition of DHA or EPA suppressed the up-regulation of elovl5 by Lxrα (P < 0.05) (Fig. 6B). Nevertheless, inhibition of Srebp-1 did not affect the regulation of elovl5 by Lxrα (P < 0.05) (Fig. 6B).

FA
Dual-luciferase reporter assays. In order to investigate the molecular mechanisms involved in the regulation of L. crocea elovl4 and elovl5, regions of 1921 (GenBank Accession No. KY863452) and 2291 bp (GenBank Accession No. KY863451), respectively, upstream of the initiation codon of the genes were cloned to perform dual-luciferase reporter assays. For elovl4, the reporter activity was 2.41-fold than the control (Fig. 7A). The croaker elovl4 reporter activity was significantly elevated by over-expression of lxrα and srebp-1 (Fig. 7A). For elovl5, the reporter activity was 1.66-fold than the control (Fig. 7B). Similarly, the L. crocea elovl5 reporter activity was significantly elevated by over-expression of lxrα (Fig. 7B). However, the over-expression of srebp-1 had no significant effect on the activity of the elovl5 promoter (Fig. 7B).

Discussion
The large yellow croaker is widely cultured in China due to its great commercial value. Recently, a study investigating the effects of FO replacement in diets for L. crocea showed that low retention of LC-PUFA in muscle occurred, thus reducing the nutritional quality of the product 26 . This was possibly due to the low LC-PUFA biosynthetic ability of large yellow croaker as generally reported for most marine species 27 . Elongases play key roles in the biosynthesis of LC-PUFA in fish 7, 11 and thus a better understanding of the underlying mechanisms regulating the transcription of elongases would contribute to strategies designed to enhance the endogenous LC-PUFA synthetic ability of the large yellow croaker that may lead to increased nutritional value of farmed products for human consumers. Seven members of the Elovl protein family have been described in vertebrates 9,28 . Typically, Elovl2 and Elovl5 play roles in the biosynthesis of LC-PUFA in vertebrates, having somewhat overlapping functions possibly related to their common evolutionary origin 29 . Whereas Elovl5 is present in virtually all teleosts, Elovl2 appears to be lost during evolution and is absent in Acanthopterygii that encompass the majority of farmed marine fish 11 . Fish generally possess two Elovl4-like elongases 11 , but only the so-called "isoform b" has PUFA as preferred substrates 25 . In the present study, we conducted the molecular cloning of an elovl4 and functionally characterized the functions of the herein reported Elovl4 and the previously cloned elovl5 20 . In addition, we investigated regulatory mechanisms of transcription factors in both elongases and the influence of "dietary" LC-PUFA in both in vivo and in   vitro systems. The results demonstrated that the expression of the two elongases was regulated by n-3 LC-PUFA via specific transcriptional factors that exert different actions on each elongase. The isolated croaker Elovl4 possessed all the features of the elongase family including transmembrane domains, a histidine box (HXXHH), and an arginine residue and single lysine residue (RXKXX) in the canonical C-terminal, indicating its role in LC-PUFA biosynthesis 30 . Zebrafish possessed two Elovl4 enzymes, Elovl4a and Elovl4b, with different functions 25 . Both the Elovl4 in zebrafish had roles in the biosynthesis of saturated VLC-FA, while Elovl4b was also able to operate on PUFA substrates producing polyenes of up to 36 carbons and thus participated in the biosynthesis of VLC-PUFA 25 . Phylogenetic analysis of the newly cloned L. crocea elovl4 showed that it was an orthologue of zebrafish Elovl4b and not Elovl4a. Consistent with these genes being orthologues of the zebrafish Elovl4b, previously cloned elovl4-like elongases isolated from cobia 26 , Atlantic salmon 31 , rabbitfish 32 and orange-spotted grouper 15 , had similar functions compared to zebrafish Elovl4b and they were all capable of elongating both saturated and polyunsaturated FA. In contrast, the L. crocea Elovl4-b like elongase did not show any activity towards endogenous saturated FA in yeast and thus does not appear to have a role in the biosynthesis of saturated VLC-FA as its previously characterized counterparts. The reasons why Elovl4 in L. crocea does not play a role in the biosynthesis of saturated VLC-FA remain uncertain, but a recent study reporting the functional characterization of Elovl4 in nibe croaker Nibea mitsukurii did not report any activity towards saturated FA using a similar expression system as that used herein. While it is possible that the activity of the nibe croaker Elovl4 towards endogenous saturated FA in yeast was simply not measured and/or reported, the close phylogenetic relationship between L. crocea and N. mitsukurii, both from the Sciaenidae family, possibly indicated that some Elovl4 from certain species or families within teleosts have subfunctionalized during evolution, as has occurred in the fatty acyl desaturase Fads2 33 , a protein family that also plays key roles in LC-PUFA biosynthesis in vertebrates 11,28 .
In agreement with the functional data obtained with fish Elovl4 15, 26, 31, 32 , the L. crocea Elovl4 showed activity towards PUFA substrates and thus confirmed its role in the biosynthesis of VLC-PUFA. Thus, the L. crocea Elovl4 exhibited high elongation efficiency towards a range of PUFA substrates, namely C 20 (20:5n-3 and 20:4n-6) and C 22 (22:5n-3 and 22:4n-6) PUFA, with C 18 substrates (18:4n-3 and 18:3n-3) being elongated to a lesser extent. The functions of VLC-PUFA are not fully understood, but studies in mammals have suggested that these compounds play pivotal roles in vision 34 and reproduction 35 . Importantly, the large yellow croaker Elovl4 could produce 24:5n-3 from directly supplemented 22:5n-3 or by step-wise elongation of 20:5n-3. Such an ability of Elovl4 has been noted as an important feature by which this enzyme can contribute to DHA biosynthesis through the Sprecher pathway in species that, like most marine farmed species including the large yellow croaker, might have lost Elovl2. Thus, the elongation product of Elovl4, namely 24:5n-3, can be desaturated to 24:6n-3 and subsequently chain shortened to 22:6n-3 by partial β-oxidation 36 Similar activities were also demonstrated for Elovl4 from zebrafish (isoform b), Atlantic salmon, cobia, rabbitfish, nibe croaker and orange-spotted grouper 15,25,31,32,37,38 .
Aquafeeds are currently formulated with decreasing levels of the marine ingredients FM and FO due to increased demand and unpredictable availability, leading to high price volatility. Nutritional regulation of genes encoding desaturases and elongases has been proposed as a possible strategy to enhance the endogenous production on n-3 LC-PUFA and optimise their retention in farmed fish 8,41,42 . In general, n-3 LC-PUFA can suppress the transcription of the Elovl4 and Elovl5 encoding genes in fish [14][15][16][17][18][19][20] . Additionally, previous studies found that LC-PUFA and their metabolites can regulate transcription of lipid metabolism related genes through modulation of transcription factors including, among others, Lxrα and Srebp-1 24 . However, the regulatory mechanisms of these transcription factors on the two herein studied elongases has been barely investigated. It has been demonstrated that n-3 LC-PUFA can suppress the transcription of the elovl4 and elovl5 elongases both in vivo and in vitro, indicating the existence of negative feedback regulation in the LC-PUFA biosynthetic pathway. Moreover, the expression of srebp-1 and lxrα were also suppressed by n-3 LC-PUFA both in vivo and in vitro in the present study, and this may largely account for the feedback suppression of expression of both elovl4 and elovl5.
In mammals, Srebp-1 positively regulates transcription via interacting with sterol response elements (SRE) in the promoter region of target genes 22 . Furthermore, Lxrα can activate transcription directly and/or indirectly via regulating Srebp-1 or other transcription factors 23,43 . Dual-luciferase reporter assays were conducted in the present study to clarify the regulatory mechanisms by which Lxrα and Srebp-1 modulate the expression of elovl4 and The results showed that the large yellow croaker Elovl5 reporter activity was induced by over-expression of lxrα, but not by over-expression of srebp-1. This may indicate that the transcription of elovl5 was up-regulated by Lxrα directly, but not dependent on Srebp-1. This observation is consistent with the results of studies carried out in Atlantic salmon and orange-spotted grouper 16,44 . The results of the in vitro experiment conducted in the present study to confirm the above findings showed that the expression of elovl5 in L. crocea hepatocytes was induced by over-expression of lxrα, while inhibition of the expression of srebp-1 had no effect on the expression of elovl5 and could not repress the activation effect of Lxrα on elovl5. This confirmed the direct stimulatory role of Lxrα on elovl5, and suggested that such regulatory mechanisms operated differently compared to mammals 24 . Promoter studies showed that the large yellow croaker Elovl4 reporter activity was induced by over-expression of both lxrα and srebp-1. This observation indicated that the expression of the L. crocea elovl4 can be up-regulated through Lxrα and Srebp-1. Additionally, the in vitro results showed that the transcription of elovl4 in hepatocyte was significantly induced by over-expression of lxrα, but significantly repressed by inhibiting the expression of srebp-1. The inhibition of the expression of srebp-1 would significantly repress the activation effect of Lxrα on elovl4. These results suggested that Lxrα can promote the expression of croaker elovl4 via a Srebp-dependent pathway. The above observations indicated differential regulation of Lxrα and Srebp-1 on Elovl4 and Elovl5 in the large yellow croaker. Knowledge of the underlying regulation mechanism of these two elongases can contribute to the development of novel practical approaches to enhance endogenous LC-PUFA production via both nutritional strategies and genetic modification.
In summary, the large yellow croaker Elovl4 exhibited all the structural features of Elovl proteins. Functional studies showed that Elovl4 and Elovl5 of L. crocea had complementary functions since the latter could effectively elongate both C 18 and C 20 PUFA substrates, while Elovl4 was more effective in the elongation of C 20

Materials and Methods
Cloning and phylogenetic analysis of the L. crocea full-length elovl4 cDNA. Total RNA was extracted from isolated croaker liver using Trizol Reagent (Takara, Japan). Thereafter, first strand cDNA was synthesized using Prime Script TM RT reagent Kit (Takara). Two degenerate primers (Elo4-F and Elo4-R) (Supplementary Table) were designed based on highly conserved regions of elovl4 cDNA sequences of other fish and used for PCR (Eppendorf Mastercycler Gradient, Eppendorf, Hamburg) using PrimeSTAR ® HS DNA Polymerase (Takara, Japan). The amplified products were purified (SK8131, Sangon Biotech, Shanghai), ligated into the pEASY-T1 vector (TransGen Biotech, China), transformed into competent Trans1-T1 cells (TransGen Biotech, China) and sequenced (Biosune Biotech, Shanghai, China), followed by blasting on GenBank to confirm its high similarity with other Elovl4 proteins. Gene specific primers (Supplementary Table) were designed to obtain the full-length elovl4 cDNA sequence by 5′-and 3′-rapid amplification of cDNA ends (RACE) PCR according to the manufacturer's instructions (SMARTer ™ RACE cDNA Amplification Kit, Clontech, CA, USA). RACE PCR products were purified, cloned into pEASY-T1 vector, and sequenced as described above.
The deduced amino acid (aa) sequence of the newly cloned croaker elovl4 cDNA was aligned with other orthologues including human (Homo sapiens, NP_073563), mouse (Mus musculus, NP_683743), rat (Rattus norvegicus, NP_001178725), zebrafish (Danio rerio, NP_957090 and NP_956266), cobia (Rachycentron canadum, ADG59898), and Atlantic salmon (Salmo salar, NP_001182481). Multiple sequence alignment was performed with Mega 6.0. A phylogenetic tree was constructed on the basis of aa sequences using the neighbor-joining method and including the newly cloned croaker Elovl4 sequence and those from other vertebrate Elovl proteins 45 . Confidence in the resulting phylogenetic tree branch topology was measured by bootstrapping the data through 1000 iterations.

Functional characterization of the L. crocea Elovl5 and Elovl4 elongases. PCR fragments cor-
responding to the open reading frames (ORF) of the Elovl5 and Elovl4 elongases were amplified from croaker liver cDNA, using High Fidelity PrimeScript ® RT-PCR Kit with primers containing restriction sites for HindIII and XhoI (Supplementary Table). The DNA fragments containing the croaker elovl5 or elovl4 ORF were digested with corresponding restriction endonucleases (Takara, Japan) and then ligated into a similarly restricted pYES2 vector (Invitrogen, UK) to yield the resulting plasmid constructs pYES2-Elo5 and pYES2-Elo4. Saccharomyces cerevisiae competent cells were transformed (S.c. EasyComp Transformation Kit, Invitrogen, USA) with the purified recombinant plasmid constructs pYES2-Elo5 or pYES2-Elo4. Transformation and selection of yeast with recombinant plasmids, and yeast culture were conducted as described in detail previously 46,47 . A single colony of transgenic yeast was grown in S. cerevisiae minimal medium −uracil supplemented with potential fatty acid (FA) substrates. For Elovl5, each culture was supplemented with one of the following substrates: 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. For Elovl4, each culture was supplemented with one of the following substrates: 18:4n-3, 18:3n-6, 20:5n-3, 20:4n-6, 22:5n-3, 22:4n-6 and 22:6n-3. In order to compensate for differences in FA update efficiency by yeast, final concentrations of FA substrates varied according to their fatty acyl chain lengths, 0.5 mM (C 18 ), 0.75 mM (C 20 ) and 1.0 mM (C 22 ) 25 . Elovl4 plays a role in the biosynthesis of very long-chain (>C 24 ) saturated FA (VLC-FA) 25 and thus pYES2-Elo4 transformed yeast were also grown in the absence of exogenously added fatty acids to enable comparison with saturated FA profiles of yeast transformed with the empty pYES2. For both Elovl5 and Elovl4 assays, yeast were grown for 2 days and then harvested, washed twice with 5 mL ice-cold HBSS (Invitrogen, UK) and freeze dried for 24 h for further analyses. Yeast transformed with empty pYES2 were additionally grown in the presence of all PUFA substrates tested in both assays for Elovl5 and Elovl4 as further control treatments.
Fatty acid analysis. Total lipids were extracted from yeast samples and fatty acids derivatized to methyl esters (FAME) as described in detail previously 48 . FAME were identified and quantified after splitless injection and run with temperature programming in an Agilent 6850 Gas Chromatograph system equipped with a Sapiens-5MS (30 m × 0.25 μm × 0.25 μm) capillary column (Teknokroma, Barcelona, Spain) coupled to a 5975 series MSD (Agilent Technologies, Santa Clara, CA, USA) as described by Monroig et al. 27 . The elongation of exogenously added PUFA substrates (18:3n-3, 18:2n-6, 18:4n-3, 18:3n-6, 18:4n-3, 20:5n-3, 20:4n-6, 22:5n-3 and 22:6n-3) was calculated by the step-wise proportion of substrate FA converted to elongated product as [areas of first product and longer chain products/(areas of all products with longer chain than substrate + substrate area)] × 100 15 . Animal experiments. The present study was carried out in strict accordance with the recommendations in the Guide for the Use of Experimental Animals of Ocean University of China. The protocols for animal care and handling, methods and experimental procedures used in this study were approved by the Institutional Animal Care and Use Committee of Ocean University of China. A 10-week feeding trial was conducted to investigate the effect of dietary n-3 LC-PUFA on expression of elovl4 and elovl5, and the regulation of certain transcription factors on the two enzymes. DHA-enriched oil (DHA content: 40.6% of total fatty acid in the form of DHA-methyl ester, Jiangsu Tiankai Biotechnology Company Limited, China) and EPA-enriched oil (EPA and DHA at 45.9% and 23.8%, respectively, of total fatty acids both in the form of TAG; Hebei Haiyuan Health Biological Science and Technology Company Limited, China) were utilized in different ratios to form three isoproteic (43% crude protein) and isolipidic (18% crude lipid) diets with low (0.46%), moderate (1.05%) and high (2.44%) n-3 LC-PUFA content (the ratio of DHA to EPA was approximately 2.0). Large yellow croaker were obtained from a local farm located in Xiangshan Bay, Ningbo, China. At the beginning of the experiment, the fish were fasted for 24 h and weighed after being anesthetized with eugenol (1:10,000) (Shanghai Reagent, China). Triplicate groups of croaker (individual weight of 10.0 ± 0.6 g) were distributed into each sea cage (1 m × 1 m × 1.5 m) at a stocking density of 60 individuals, and fed twice daily (05:00 and 17:00 h) to apparent satiation for 70 days. At the end of the experiment, croaker were fasted for 24 h and anesthetized and culled by an overdose of MS222 (ethyl 3-aminobenzoate methanesulfonic acid salt, Aldrich, USA). Livers of three fish per cage were isolated, frozen in liquid nitrogen and stored at −80 °C for subsequent gene expression analysis of elongases (elovl4 and elovl5) and transcription factors (lxrα and srebp-1). In addition, various tissues (eye, brain, testis, heart, liver, kidney, stomach, intestine and muscle) were isolated from nine croaker individuals fed diets with moderate n-3 LC-PUFA levels (1.05%) to investigate the tissue distribution of elovl4 mRNA.
Primary hepatocyte isolation and incubation. The above feeding study provided preliminary data on the in vivo nutritional regulation of elovl4 and elovl5. To confirm and support these data, an in vitro study was conducted to further investigate the role of certain transcription factors on the regulation of elovl4 and elovl5 elongases in L. crocea. Hepatocytes were isolated from five yellow croaker individuals (~50 g) starved for 24 h according to the published protocols 49 with slight modification 50,51 . Briefly, croaker were anesthetized with MS 222 and the branchial arch cut followed by immersion in 70% ethanol for 3 min to sterilize the external surface. Liver tissues were excised aseptically and rinsed twice with phosphate buffered saline (PBS, pH 7.4 at 4 °C) supplemented with amphotericin-B (25 μg mL −1 ), streptomycin (100 μg mL −1 ) and penicillin (100 IU mL −1 ). Thereafter, the liver was aseptically minced into 1 mm 3 pieces, followed by digestion in 0.25% sterile trypsin at room temperature for 15 min. Trypsin was neutralized with Dulbecco's Modified Eagle Medium (DMEM, Gibco, USA) containing 20% fetal bovine plasma (FBS, Invitrogen, USA). The cell suspension was collected and filtered through sterile 75 μm mesh, followed by centrifugation at low speed (100 g, 5 min) and the supernatant discarded. Isolated cells were washed in red blood cell lysis buffer (Beyotime Institute of Biotechnology, Haimen, China) for 2 min at 4 °C. Cell viability was evaluated using a hemocytometer under an inverted microscope after the cells were stained with 0.4% Trypan Blue. The hepatocytes were re-suspended in DMEM medium containing 1 mM glutamine, 20% FBS, penicillin (100 IU mL −1 ) and streptomycin (100 μg mL −1 ). Cells suspensions with more than 95% cell viability were used for the subsequent experiments.
Croaker primary hepatocyte suspensions were incubated with fatty acid (DHA or EPA) to confirm the influence of n-3 LC-PUFA on the transcription of the elongases (elovl4 and elovl5) and transcription factors (lxrα and srebp-1) in vivo. Fatty acid (DHA and EPA, Cayman Chemical Co., USA) was supplemented to cells in the form of BSA/fatty acid complexes that were prepared at 10 mM concentration according to Ou et al. 52 and stored at −20 °C. Additionally, inhibitors/agonists of transcription factors were used to clarify the role of the transcription factors on the regulation of elovl4 and elovl5 elongases in L. crocea. GW3965 HCl (Selleckchem, Shanghai, China) was used as an LXRα agonist whereas FGH10019 (MCE, USA) was used as a SREBP-1 inhibitor, respectively. Cells were seeded in 6-well plates with a density of 2 × 10 6 viable cells per well in DMEM/F12 (Gibco) containing 20% FBS, 100 U ml −1 penicillin and 100 μg ml −1 streptomycin, followed by incubation for 24 h. The hepatocytes were then washed and incubated for 1 h in FBS-free DMEM/F12 medium prior to incubation with EPA, DHA and the above inhibitors or agonists in triplicate wells. After incubation, cells were lysed in the wells and harvested for RNA extraction.
Real-time quantitative PCR (RT-qPCR) analysis. Specific primers for RT-qPCR were designed by Primer Premier 5.0 (Premier Biosoft) based on the cloned nucleotide sequences. The stability of β-actin was verified and confirmed and used as the reference gene 53 . The amplification was performed in a quantitative thermal cycler (Mastercycler EP Realplex, Eppendorf, Germany). The program was 95 °C for 2 min, followed by 40 cycles of 95 °C for 10 s, 57 °C for 10 s, and 72 °C for 20 s. At the end of each reaction, melting curve analysis of amplification products was carried out to confirm that a single PCR product was present in these reactions. The amplification efficiencies of the target and reference genes were determined from the slope of the log-linear portion of the calibration curve and PCR efficiency = 10 (−1/Slope) − 1. The expression levels of the target genes were calculated following the 2 −ΔΔCt method described by Livak & Schmittgen 54 . Cloning of the elovl4 and elovl5 promoters. Genomic DNA was extracted from croaker liver using the SQ Tissue DNA Kit (OMEGA, USA) according to the manufacturer's instructions. Then, the genomic DNA was digested with four restriction enzymes (DraI, EcoRV, PvuII and StuI), purified and ligated to Genome Walker Adaptors following the instructions of Universal Genome Walker 2.0 Kit user manual (Clontech, USA). The promoter region of elovl5 and elovl4 was amplified through a nested (two round) PCR combining the kit supplied primers Ap1 and Ap2 with Advantage 2 PCR Kit (Clontech, USA) and the gene specific primers designed on elovl5 and elovl4 cDNA (see Supplementary Table for primer details). The conditions for the primary round PCR were: 7 cycles of 25 s at 94 °C, 3 min at 72 °C, followed by 32 cycles of 25 s at 94 °C, 3 min at 65 °C, then with an additional 7 min at 67 °C after the final cycle. The conditions for the secondary round PCR were: 5 cycles of 25 s at 94 °C, 3 min at 72 °C, followed by 20 cycles of 25 s at 94 °C, 3 min at 65 °C, then with an additional 7 min at 67 °C after the final cycle. The PCR products were purified, cloned into pEASY-T1 and sequenced as described above.
Expression and reporter plasmids constructs. For expression plasmids, PCR fragments corresponding to the ORF of the croaker srebp-1 (GenBank accession number: KP342262) were amplified with primers containing restriction sites for EcoRI and XhoI, respectively, and the ORF of croaker lxrα (GenBank accession number: XM_019273432) was amplified with primers containing restriction sites for ClaI and EcoRI (Supplementary Table). The DNA fragments were digested with corresponding restriction endonucleases (Takara) and ligated into a similarly restricted pCS2 + vector (Invitrogen, USA) to yield the expression plasmid constructs pCS2-srebp1and pCS-lxrα. The reporter plasmid Elovl5-Luc and Elovl4-Luc were engineered to contain a fragment of the large yellow croaker Elovl5 and Elovl4 promoters cloned into the KpnI/XhoI sites of the pGL3 basic vector (Promega, USA) containing the coding sequences of firefly luciferase cDNA. The renilla luciferase plasmid pRLTK (Promega) was used as an internal control. Plasmids for transfection were prepared using TransGen Plasmid Mini Kit (Beijing, China) according to the manufacturer's instructions.

Cell culturing, transfection and luciferase assay. HEK 293 T cells were cultured in DMEM (Gibco)
supplemented with 10% fetal bovine plasma (Invitrogen) at 37 °C in a humidified incubator under 5% CO 2 . For DNA transfection, cells were seeded in 24-well plates until they were 90-100% confluent at the time of transfection. Then plasmids were transfected by using the Lipofectamine TM 2000 Reagent (Invitrogen) according to the manufacturer's instruction. Briefly, 0.6 μg expression plasmid, 0.2 μg reporter gene plasmid, 0.02 μg pRLTK renilla luciferase plasmid, and 2.0 μL Lipofectamine TM 2000 were co-transfected into the cells in each well in the 24-well plate. All assays were performed with three independent transfections. Firefly and renilla luciferase activities were measured using Dual Luciferase Reporter Assay System (Promega, USA) according to the manufacturer's instructions. Briefly, after 48 h transfection, HEK293T cells in the 24-well plates were washed twice with 100 μL PBS, then lysed with 100 μL 1x passive lysis buffer (PLB) at room temperature for 10 min. Cell lysate (20 μL) was transferred to a clean plate and 50 μL of luciferase assay reagent II and 1x Stop & Glo substrate were added in sequence, then firefly and renilla luciferase activities were measured using a InfiniTE200 plate reader (Tecan, Switzerland).

Statistical analysis.
The results were presented as means ± standard error of the mean (SEM). Data from each treatment were subjected to one-way analysis of variance (ANOVA) and correlation analysis where appropriate using SPSS 19.0 for Windows. Tukey's multiple range test was chosen as a multiple comparison test with a significance level of 0.05. For the Elovl4 functional characterization, the saturated VLC-FA profiles from yeast expressing the elovl4 were compared to the control by a Student's t-test (P < 0.05).