Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Parental high dietary arachidonic acid levels modulated the hepatic transcriptome of adult zebrafish (Danio rerio) progeny

  • Anne-Catrin Adam ,

    Roles Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing – original draft

    Affiliation Institute of Marine Research, Nordnes, Bergen, Norway

  • Kaja Helvik Skjærven,

    Roles Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Writing – review & editing

    Affiliation Institute of Marine Research, Nordnes, Bergen, Norway

  • Paul Whatmore,

    Roles Data curation, Formal analysis, Validation, Visualization, Writing – review & editing

    Affiliation Institute of Marine Research, Nordnes, Bergen, Norway

  • Mari Moren,

    Roles Conceptualization, Funding acquisition, Investigation, Writing – review & editing

    Current address: Nofima AS, Bergen, Norway

    Affiliation Institute of Marine Research, Nordnes, Bergen, Norway

  • Kai Kristoffer Lie

    Roles Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Writing – review & editing

    Affiliation Institute of Marine Research, Nordnes, Bergen, Norway

Parental high dietary arachidonic acid levels modulated the hepatic transcriptome of adult zebrafish (Danio rerio) progeny

  • Anne-Catrin Adam, 
  • Kaja Helvik Skjærven, 
  • Paul Whatmore, 
  • Mari Moren, 
  • Kai Kristoffer Lie


Disproportionate high intake of n-6 polyunsaturated fatty acids (PUFAs) in the diet is considered as a major human health concern. The present study examines changes in the hepatic gene expression pattern of adult male zebrafish progeny associated with high levels of the n-6 PUFA arachidonic acid (ARA) in the parental diet. The parental generation (F0) was fed a diet which was either low (control) or high in ARA (high ARA). Progenies of both groups (F1) were given the control diet. No differences in body weight were found between the diet groups within adult stages of either F0 or F1 generation. Few differentially expressed genes were observed between the two dietary groups in the F0 in contrast to the F1 generation. Several links were found between the previous metabolic analysis of the parental fish and the gene expression analysis in their adult progeny. Main gene expression differences in the progeny were observed related to lipid and retinoid metabolism by PPARα/RXRα playing a central role in mediating changes to lipid and long-chain fatty acid metabolism. The enrichment of genes involved in β-oxidation observed in the progeny, corresponded to the increase in peroxisomal β-oxidative degradation of long-chain fatty acids in the parental fish metabolomics data. Similar links between the F0 and F1 generation were identified for the methionine cycle and transsulfuration pathway in the high ARA group. In addition, estrogen signalling was found to be affected by parental high dietary ARA levels, where gene expression was opposite directed in F1 compared to F0. This study shows that the dietary n-3/n-6 PUFA ratio can alter gene expression patterns in the adult progeny. Whether the effect is mediated by permanent epigenetic mechanisms regulating gene expression in developing gametes needs to be further investigated.


In today’s dietary pattern, we observe a selective decrease of n-3 polyunsaturated fatty acids (PUFAs) in favour of n-6 PUFAs. This results in a decreased n-3/n-6 PUFA ratio [1, 2]. Physiological effects of a decreasing n-3/n-6 PUFA ratio are diverse, but taken together, studies indicate that a disproportional high intake of n-6 PUFAs may contribute to health problems [37]. In the past decades, several studies have shown the benefit of increased n-3 PUFA levels in the diet [8, 9], and focus has been directed on n-3 PUFAs such as eicosapentaenoic acid (EPA, 20:5n-3) and docosahexaenoic acid (DHA, 22:6n-3). Arachidonic acid (ARA, 20:4n-6), an n-6 PUFA that competes for the same enzymes and therefore is interlinked, has been less studied. ARA and its derivatives, have vital roles in growth and various signalling cascades regulating inflammatory processes, bone metabolism and reproduction as observed in different species [1014]. n-3 and n-6 PUFAs have the potential to change cellular phenotypes by changing membrane lipid composition and controlling gene expression through activating nuclear receptors [1517]. In addition, n-3 and n-6 PUFAs can also affect DNA methylation patterns [1820]. Recently, we demonstrated that high dietary ARA levels fed to zebrafish affected the levels of oxidized amino acids and lipids, and changed the immune-related eicosanoids and lipid metabolism [21]. More and more studies indicate that diet also affect following generations in terms of long-term health of the progeny [2226]. Here, we investigate whether high ARA given to the parents’ generation can impact the progeny’s transcriptome. However, little is known about how changes in the parental dietary n-3 and n-6 PUFA composition impact the adult progeny.

The period of oocyte and spermatozoa maturation displays a sensitive window, where parental nutrition has metabolic influence on future fertilized eggs [27]. Another way that parental diet effects can be mediated is through transcripts deposited in the newly fertilized egg that regulate early embryonic development and thus determine future gene expression patterns, growth and physiology [2831]. It has been shown in zebrafish that dietary micronutrient status of the parents can influence gene expression patterns of their embryos and livers of their adult offspring [3234]. Furthermore, nutritional induced obesity of the parents has shown to affect fertility (egg production) and gene expression of zebrafish eggs [35]. Studies on different vertebrate species have demonstrated that the maternal dietary n-3 and n-6 PUFA profile influenced oocyte composition, embryonic development and health of progeny [9, 3639]. In teleost, dietary n-3 and n-6 PUFA composition was found to affect oocytes and reproductive performance [4044], but little is known about the changes in gene expression profiles in adult progeny.

Zebrafish (Danio rerio), a tropical freshwater teleost fish, is an acknowledged vertebrate model organism. It has been widely used in research to increase our understanding of gene function and the importance of nutrition in outcomes related to development, health and disease in vertebrates [4552]. In the present study, we fed parental zebrafish either a control diet (low in ARA) or a diet high in ARA, whereas progeny from both dietary groups were fed the control diet until adulthood. We wanted to investigate the impact of parental high dietary ARA levels on transcriptomic patterns in adult progeny.

Materials and methods

Ethical considerations

This zebrafish feeding trial was conducted in accordance with the Norwegian Animal Research Authority and approved by the Norwegian Food Safety Authority (division no. 54, reference 2012/145126) according to the current Regulation on Animal Experimentation (FOR 1996-01-15 no. 23).

Feeding trial and zebrafish husbandry

Standardized operating procedures for mating, handling and feeding for both F0 and F1 generation of wildtype AB zebrafish (Danio rerio) has previously been reported [32]. Briefly, F0 embryos were collected randomly and larvae were fed with Gemma micro® (Skretting, Stavanger, Norway) as a start feed from 5 days post fertilization (DPF) and Artemia nauplii (Silver Star Artemia, Salt Lake, USA) from 7 DPF until 26 DPF (Fig 1). The experimental diets were given twice a day from 27 DPF onwards. Control and high ARA diet composition can be found in S1 File [21]. Progeny (F1 generation), from both parental diet groups, were fed as the F0 control fish with the experimental control diet from 27 DPF until sampling. Fish were kept in 10 gender mixed tanks (containing 60 fish each until 44 DPF and thereafter reduced to 20 fish each) per diet group. All fish were kept under steadily monitored standard conditions with 28±1°C, 14 h light-10 h dark period, conductivity of 500 μS, 6 ppm (mg/L) dissolved oxygen and pH 7.5 in tanks in a reverse osmosis water treatment system (Aquatic Habitats® recirculation system, MBKI Ltd, Calverton, GBR). F0 generation was mated at 97 DPF.

Fig 1. Experimental design of the transgenerational zebrafish feeding trial.

F0 zebrafish from both control and high ARA group were fed a start feed containing Gemma micro® and Artemia nauplii from 5 DPF until 26 DPF. The two experimental groups were given either a control or high ARA diet from 27 DPF onwards until sampling. F0 fish were mated at 97 DPF to produce F1 generation. Both groups in the F1 generation were fed the control diet from 27 DPF until sampling. F0 and F1 body weight (grams) and liver tissue sampling for transcriptome analysis (RNA-sequencing) were performed at 154–156 DPF (F0) and 140–142 DPF (F1).

Liver sampling and RNA extraction

Prior to dissection, fish were deprived of food for 18 h, anesthetized with 0.05% Tricaine Methane Sulphonate (PHARMAQ AS, Oslo, Norway), blotted dry on tissue paper prior to weighing, euthanized by cutting the cardinal vein and the liver was dissected subsequently. Livers were sampled in random order between 154–156 DPF (F0) and 140–142 DPF (F1) due to simultaneous sampling for other analyses connected to this trial. Six biological replicates representing six different tanks for each of the dietary groups, of which each replicate is a pool of six male livers from one tank. Livers were briefly rinsed in 1x PBS, snap frozen with liquid nitrogen and stored at -80°C for transcriptome analysis (RNA-sequencing). Total RNA was extracted using QIAzol Lysis Reagent (Qiagen, Hilden, Germany) and RNA samples were DNase treated with the AmbionTM DNA-freeTM DNA Removal Kit (Invitrogen, Thermo Fisher Scientific, Waltham, USA) in order to avoid remaining genomic DNA. RNA quantity was verified using NanoDrop® ND-1000 Spectrophotometer (NanoDrop Technologies, Wilmington, USA). RNA integrity (RIN) was determined using an Agilent 2100 Bioanalyser (RNA 6000 Nano LabChip kit, Agilent Technologies, Santa Clara, USA). RIN values were on average 9.06±0.39.

RNA high-throughput sequencing and data processing

The Norwegian Sequencing Centre (NSC) performed RNA-sequencing (RNA-seq) and library preparation using TruSeqTM Stranded mRNA Library Prep Kit (Illumina, Inc, San Diego, USA). Libraries were sequenced on the NextSeq500 platform (Illumina, Inc, San Diego, USA) to generate single-end 75bp reads. Sequence quality was assessed using FastQC v0.11.5. Finding high quality (Phred scores almost universally above 30) and close to zero adapter contamination on the raw reads, we decided that mapping untrimmed reads to the genome was the optimal strategy, instead allowing the mapping software to exclude errors through discarded mismatches [53]. An average of 10 047 201 reads per sample were mapped to the GRCz10 (Genome Reference Consortium Zebrafish Build 10) assembly based on both RefSeq (GCF 000002035.5 GRCz10) and Ensembl [54] using the default parameters of HISAT2 [55] resulting in an average of 76.29% of reads unambiguously assigned to RefSeq genes and 82.22% of reads unambiguously assigned to Ensembl genes. Read counts per gene were quantified using featureCounts [56] and pre-filtered to exclude combined mean read counts smaller than 10.

Bioinformatic analysis

Differential gene expression was estimated using DESeq2 [57]. By default, internal normalization was performed to correct for variable sequencing depth and library size. Wald-test was used for significance testing and Benjamini-Hochberg for p-value false discovery correction (adjusted p). DESeq2 analysis and visualisation of data were performed in R ( Mapping against different reference genomes can produce variable expression values and differentially expressed genes (DEGs) identified [58]. The annotated DEG lists from both RefSeq and Ensembl reference genomes are given in S1S8 Tables. However, enrichment and downstream analyses were based on concordant DEGs between both reference genome annotations (S9S12 Tables). The data discussed in this publication have been deposited in NCBI's Gene Expression Omnibus [59] and are accessible through GEO Series accession number GSE104692 (

Concordant gene symbols of DEGs (adjusted p-value<0.05) of F1 generation were sent to Ingenuity® Pathway Analysis software suite (IPA®, Ingenuity Systems, Qiagen, Redwood City, USA) for downstream analysis. DEGs with corresponding adjusted p-value and log2 fold change (log2FC) were imported into IPA® as human orthologues (S13 Table). An overlap p-value (right-tailed Fisher`s Exact test, p<0.05) and an activation z-score for the correlation between the imported RNA-seq dataset and the Ingenuity® Knowledge Base is calculated. IPA® integrates direct DEG changes to predict an upregulation or downregulation of canonical pathways and biological functions in F1 high ARA livers using z-scores. DEGs from the comparison of both control groups and both high ARA groups between the generations were used for functional annotation for KEGG pathways and GO terms by over-representation testing using the R package ‘clusterProfiler’ [60] (S14 Table).

Validation of RNA-sequencing by RT-qPCR

Reverse transcription followed by quantitative real-time PCR (RT-qPCR) was performed as previously described [61] for validating the RNA-seq data. Reverse transcription and PCR of standard curve and individual samples was run with the Gene Amp 9700 PCR machine (Applied Biosystems, Foster City, USA). Real-time RT-qPCR was performed starting with a 5 min template incubation and denaturation step at 95°C, followed by 45 cycles divided in 10 s denaturation at 95°C, 10 s annealing at 60°C and 10 s synthesis at 72°C using the CFX384 Touch™ Real-Time PCR Detection System (Bio-Rad Laboratories, Inc., Hercules, USA) with the LightCycler® 480 SYBR Green I Master kit (Roche Applied Science, Penzberg, Germany). Samples were amplified in triplicates and the mean was used for further calculations. Normalised expression of target genes was determined using the geNorm algorithm [62] based on the geometric mean of 3 reference genes: eef1a1I1 (Eukaryotic translation elongation factor 1 alpha 1, like 1) [34], tuba1c (Tubulin, alpha 1c) [63] and actb1 (Actin, beta 1) (S2 File). We investigated gene transcription for fasn (Fatty acid synthase), mat1a (Methionine adenosyltransferase I, alpha), cbsb (Cystathionine-beta-synthase b) and vtg5 (Vitellogenin 5) that were selected among the differentially expressed genes in F1 livers from the RNA-seq analysis.

Statistical analysis

Differences in body weight between the dietary groups are presented as mean with standard deviation (SD) and an unpaired, two-tailed t-test was used for significance testing (p-value<0.05). For validation of the RNA-seq results by real-time RT-qPCR, gene expression was tested for group differences using an unpaired, two-tailed, non-parametric t-test (S2 File). Statistical significance analysis of F0 and F1 body mass and RT-qPCR gene expression results was performed with GraphPad Prism 6 software (GraphPad Software, Inc, San Diego, USA).


Body weight of F0 and F1 zebrafish

We observed no changes in body weight between the diet groups in both F0 and F1 generation (Table 1).

Liver gene expression patterns

Principal component analysis (Fig 2) and volcano plot comparison (Fig 3) showed a clear separation between the dietary groups in F1. This was not as clear for the F0 generation. It also shows a clear separation based on generation (F0 vs F1), even between the control groups.

Fig 2. Principal component analysis of RNA-sequencing data from male livers in F0 generation fed either a control or high ARA diet and their F1 progeny fed the control diet.

The plot displays high similarity in the transcriptome of F0 control and high ARA livers compared to clearer separation in gene expression patterns between the F1 diet groups. The magnitude of variation between replicates within a diet group were similar among all groups. Plot shows data underlying log-transformed read counts based on RefSeq reference genome mapping.

Fig 3. Volcano plot of RNA-sequencing data from male livers in F0 generation fed either a control or high ARA diet and their F1 progeny where both groups received the control diet.

Presented data represents overlapping genes from both RefSeq and Ensembl reference genome mapping (GRCz10). Red spots represent differentially expressed genes (DEGs) between control and high ARA group in F0 (A) and F1 (B) generation. The green line denoted the significance threshold (adjusted p<0.05) for DEGs.

We found 20 DEGs between control and high ARA (adjusted p<0.05) in F0 generation (Fig 3A and S3 File) and 470 DEGs in F1 generation (Fig 3B and S3 File). Comparing F0 and F1 generation, 428 and 1987 DEGs were found to be differentially expressed between F0 and F1 control (S3 and S4 File) and between F0 and F1 high ARA groups (S3 and S4 Files), respectively.

ARA induced modulation of the parental (F0) liver transcriptome.

No functional enrichment analysis was performed due to low DEG count in F0 generation. However, among the 20 DEGs (S9 Table), we found two affected genes relevant for lipid metabolism: ncoa3 (nuclear receptor coactivator 3) involved in co-activation of different nuclear receptors like retinoid x receptors (RXRs) and peroxisome proliferator-activated receptors (PPARs), and bbox1 (gamma-butyrobetaine hydroxylase 1) involved in the biosynthesis of carnitine, which is essential for fatty acid supply in β-oxidation.

ARA induced modulation of the progeny (F1) liver transcriptome.

DEGs (adjusted p<0.05) from F1 generation were sent to downstream analysis using IPA® (S13 Table). Parental high ARA diet was associated with differential expression of a diverse set of genes clustering in several canonical pathways of which ‘eIF2 signalling pathway’ was predicted to be the most significantly downregulated pathway (Table 2).

Table 2. Significant canonical pathways associated with DEGs from F1 high ARA livers proposed by IPA®.

Among the lipid metabolism related biological functions, the top most significantly enriched functions are shown in Table 3. The full list is given in S5 File where various biological functions related to phospholipid, steroid, long chain fatty acid and cholesterol metabolism were enriched by the F1 DEGs. ‘LXR/RXR activation’ canonical pathway is predicted to be downregulated (z = -2; Table 2). ‘PPARα/RXRα Activation’ was found to be a significantly enriched canonical pathway in the high ARA group. Pparaa showed higher expression levels (p = 0.005, S4 Table) as shown in Fig 4. ‘Oxidation of fatty acids’ is one of the top most enriched biological functions (Table 3). For acox1 being involved in the first enzymatic step during peroxisomal β-oxidation and acting downstream of PPARα, showed upregulated expression in high ARA livers (Fig 4). Helz2 which encodes a nuclear transcriptional co-activator for PPARα, was downregulated. High ARA livers showed an upregulated expression of genes involved in the long-chain fatty acid biosynthesis (acaca, fasn and srebf1) compared to livers arriving from control group (Fig 4). Elovl4b, which is involved in very long-chain fatty acid elongation was found to be significantly downregulated in high ARA livers. We found an upregulated gene expression for dagla that is involved in the synthesis of 2-arachidonoyl-glycerol, an endocannabinoid. Genes like prkcq and dgkza play roles in lipid signalling pathways like T cell receptor signalling and showed higher gene expression levels in high ARA group compared to control group.

Fig 4. Differential expression of genes involved in retinoid processing and putative interactions with the PPARα/RXRα pathway and lipid signalling in male livers that are associated with parental high dietary ARA levels.

Filled shapes in the figure represent overlapping differentially expressed genes between RefSeq and Ensembl annotation. Shapes that are highlighted with glow underlie gene expression information exclusively from Ensembl annotation (S4 Table); acaca (acetyl-CoA carboxylase alpha), aclya (ATP citrate lyase a); acox1 (acyl-CoA oxidase 1, palmitoyl; alias: peroxisomal acyl-CoA oxidase 1); aldh1a2 (aldehyde dehydrogenase 1 family, member A2); bco1 (beta-carotene oxygenase 1); crabp2b (cellular retinoic acid binding protein 2, b); dagla (diacylglycerol lipase, alpha); dgkza (diacylglycerol kinase, zeta a), elovl4b (ELOVL fatty acid elongase 4b); fabp7a (fatty acid binding protein 7, brain, a); fasn (fatty acid synthase); helz2 (helicase with zinc finger 2, transcriptional coactivator); lrata (lecithin retinol acyltransferase a (phosphatidylcholine-retinol O-acyltransferase)); lratb (lecithin retinol acyltransferase b (phosphatidylcholine-retinol O-acyltransferase)); pparaa (peroxisome proliferator-activated receptor alpha a); prkcq (protein kinase C, theta); rbp2b (retinol binding protein 2b, cellular); rdh1 (retinol dehydrogenase 1); retsat (retinol saturase (all-trans-retinol 13,14-reductase)); rxraa (retinoid X receptor, alpha a); scarb1 (scavenger receptor class B, member 1); srebf1 (sterol regulatory element binding transcription factor 1); vldlr (very low density lipoprotein receptor).

Table 3. The top most significantly enriched lipid metabolism related biological functions associated with DEGs from F1 high ARA livers proposed by IPA®.

‘Metabolism of retinoid’ was among the enriched lipid metabolism related biological functions for F1 DEGs (S5 File). Transcripts of genes encoding enzymes involved in retinaldehyde synthesis from beta-carotene (bco1) and subsequent synthesis to retinoic acid (aldh1a2) as well as retinol saturation (retsat) were found to be downregulated in the high ARA group (Fig 4). Rxraa that is transcriptionally regulated by stereoisomers of retinoic acid, was shown to be downregulated in livers associated with parental high ARA diet. Two transcripts encoding transporters for beta-carotene (scarb1) and retinol (rbp2b) were downregulated.

Among other enriched canonical pathways (Table 2), we found significantly upregulated transcripts of mat1a, prmt1, cbsb, cth and got1 that clustered in ‘cysteine biosynthesis III (mammalian)’, ‘cysteine biosynthesis/homocysteine degradation’ and ‘superpathway of methionine degradation’ (Fig 5). Cysteine is provided through the transsulfuration pathway (cbcb, cth) for glutathione metabolism where gsta.1 involved in glutathione detoxification was found to be downregulated in the high ARA group. gls2b and glud1a that are related to glutamate metabolism showed decreased expression in high ARA livers (Fig 5). aldh1l1, which is involved in the folate cycle, was downregulated in high ARA livers compared to control livers.

Fig 5. Parental high ARA levels are associated with differential expression of genes involved in methionine cycle, transsulfuration pathway, and glutamate and glutathione metabolism in male F1 livers.

Filled shapes in the figure represent overlapping differentially expressed genes between RefSeq and Ensembl annotation. Shapes that are highlighted with glow underlie gene expression information exclusively from Ensembl annotation (S4 Table); ahcy (adenosylhomocysteinase); aldh1l1 (aldehyde dehydrogenase 1 family, member L1); cbsb (cystathionine-beta-synthase b); cth (cystathionase (cystathionine gamma-lyase)); gart (phosphoribosylglycinamide formyltransferase); gclc (glutamate-cysteine ligase, catalytic subunit); gls2b (glutaminase 2b (liver, mitochondrial); glud1a (glutamate dehydrogenase 1a); got1 (glutamic-oxaloacetic transaminase 1); gpx1b (glutathione peroxidase 1b); GSH (reduced glutathione); gsr (glutathione-disulfide reductase); gss (glutathione synthetase); GSSG (oxidized glutathione); gsta.1 (glutathione S-transferase, alpha tandem duplicate 1); mat1a (methionine adenosyltransferase I alpha); msra (methionine sulfoxide reductase A); mthfr (methylenetetrahydrofolate reductase (NAD(P)H)); mtr (5-methyltetrahydrofolate-homocysteine methyltransferase); oplah (5-oxoprolinase (ATP-hydrolysing)); prmt1 (protein arginine methyltransferase 1); R-S-glutathione (glutathione-S-conjugate); SAH (S-adenosyl-homocysteine); SAM (S-adenosyl-methionine); γ-Glu-Cys (gamma-glutamyl-cysteine); THF (tetrahydrofolate); 5-methyl-THF (5-methyltetrahydrofolate); 10-COH-THF (10-formyl-tetrahydrofolate); 5,10-meTHF (5,10-methylene-THF).

We observed increased gene expression of the estrogen receptor 1 (esr1) in livers associated with parental high ARA levels. Steroid metabolism related functions such as ‘concentration of progesteron’, ‘metabolism of estrogen’, ‘sulfation of beta-estradiol’ were suggested to be enriched by IPA® (Table 3 and S5 File). Among the unmapped IDs, vitellogenin 5 (vtg5, no human orthologue) expression was found to be upregulated in high ARA livers compared to the control. In F0 high ARA livers, both vtg5 (p = 0.01) and esr1 (p = 0.02) showed decreased expression (S2 Table), but did not meet the chosen threshold requirements (adjusted p<0.05) for downstream analysis.

None of the F1 DEGs were in matching with those in F0 generation due to the chosen cut-off threshold.

Intergenerational differential gene expression.

7.9% of the DEGs were overlapping between the intergenerational control and high ARA group comparison (S3 File). Functional annotation for KEGG pathways and GO terms of DEGs between both F0 and F1 control groups, and between F0 and F1 high ARA groups are given in S14 Table. Comparing the control groups, significantly enriched KEGG pathways and GO terms (q-value cut-off <0.05) related to amino acid biosynthesis and oxidoreduction related processes among others were identified. For the high ARA groups, mainly transcript processing, translation and protein biosynthesis related functions were identified.

Confirmatory RT-qPCR for RNA-sequencing validation.

Expression of fasn, vtg5, cbsb and mat1a were investigated in the F1 livers by real-time RT-qPCR for RNA-seq validation purposes. RT-qPCR and RNA-seq derived expression patterns of those target genes were similar comparing means of control and high ARA replicates (S2 File).


In the present study we identified changes in hepatic gene expression patterns in the adult zebrafish progeny of parental fish given high dietary ARA levels. Despite the large mRNA expression differences observed in F1 generation, surprisingly few DEGs were found in F0 generation. The parental fish (F0) that were given the high ARA or control diet for 17 days (44 DPF) showed major metabolic profile differences as investigated in our previous article [21]. Thus it is difficult to explain the low effect at gene expression level in the parental generation. Variation in gene expression is often invoked to explain metabolic differences [64]. In addition, for this study we found no overlap between the two generations, and even reducing the cut-off stringency for DEGs in F0 generation gave very few genes overlapping between F0 and F1 generation. Furthermore, the differences between F0 control and F1 control patterns were larger than expected. One would expect that they cluster closer together as both were fed the control diet. Our data indicates that intergenerational differences in gene expression are greater than the intragenerational differences between the dietary groups. Although many factors might apply, it is conceivable that varying age of the fish as well as individual differences between the generations can give differences. Possibly, a differently composed diet fed to previous generations of the present F0 generation might have influenced F0 transcriptomic patterns. The latter one would also explain the small differential gene expression between F0 control and high ARA group. We can also not exclude the possibility of introducing bias due to different sampling points. Since it is difficult to identify the factors causing the intergenerational differences, we are obliged to be cautious interpreting the differences with regards to changes in pathways. Future studies should be conducted to reveal the significance of these results. The intergenerational analysis can be viewed in the supplementary files.

When comparing body weight in F0 and F1 generation, we previously reported a difference in F0 juveniles, whereas this effect disappeared at later stages (91 DPF) [21]. Similar, there were no differences in body weight between the two groups of adult F1 progeny.

Despite the weak coherence between transcriptomic and metabolic profile in F0, there was a link between the parental metabolic data and the gene expression in the F1 progeny. In the previous metabolomics study of the F0 parental fish, we observed decreased levels of glutathione, glutamine and cysteine and increased levels of oxidized metabolites of cysteine and methionine derivatives indicating an oxidized environment due to increased n-6 PUFA levels. Similarly in the present study, transcriptomic analysis of progeny livers (F1 high ARA) implied an increased expression of genes involved in the methionine cycle, cysteine (transsulfuration) and glutamate synthesis (Fig 5). This can in turn involve alterations in pathways such as glutathione or folate-mediated 1-C metabolism by affecting redox homeostasis or methylation-dependent functions such as for epigenetic modifications. The findings described above indicate major compensatory mechanisms in the progeny most likely owing a modulation of the transcriptome by the oxidized and pro-inflammatory environment previously reported in the parental fish fed high ARA. Despite giving both progeny groups the control diet, we found major differences in hepatic transcriptomic profiles at adult stage.

The present results suggest that parental high dietary ARA affected RXR activated pathways in the progeny. ‘PPARα/RXRα Activation’ and ‘LXR/RXR Activation’ pathways were among the top affected pathways according to the IPA® analysis. Several genes belonging to retinol metabolic pathway were differentially expressed in adult progeny. Retinoids and their metabolites are potent activators controlling a range of essential physiological processes such as growth, limb patterning, eye vision, spermatogenesis and cell differentiation [65, 66]. Regulating their action is important for normal embryo development and epithelial differentiation, and disruption of signalling can have detrimental effects on the organism [6770].

Retinoid metabolites act on lipid signalling pathways by activating RXRs which in turn form heterodimers with PPAR-lipid complexes (Fig 4). The PPAR complex controls transcription of target genes involved in lipid signalling and metabolism [7173]. Several genes acting downstream of the nuclear receptor PPARα/RXRα complex were also found to be differently expressed in F1 high ARA livers. Genes, which encode enzymes regulating fatty acid levels, particularly by influencing fatty acid biosynthesis, transport and peroxisomal β-oxidation. Modulating these pathways influence energy expenditure, membrane composition and fatty acid distribution. In addition, oxidation of fatty acids were among the top lipid metabolic pathways enriched in the F1 high ARA group. We reported previously that elongated ARA metabolites and dicarboxylic acid levels were increased in the parental high ARA group [21]. Those results suggested an increased β-oxidation, particularly peroxisomal β-oxidative degradation of long-chain fatty acids, which can be associated with the observed transcriptomic changes related to fatty acid metabolism in the progeny.

Furthermore, studies have also shown a link between lipid and folate metabolism where PPARα seem to be involved in the regulation of key enzymes along the choline oxidative pathway which is closely linked to the methionine cycle [74, 75]. In the present study IPA® reported the methionine degradation pathway as one of the significantly affected pathways in the progeny, despite no indicated direction of influence (z-score). It has been previously shown that the expression of genes regulating homocysteine synthesis from methionine were sensitive to either high dietary n-3 or n-6 PUFAs [76]. It is conceivable that our above described gene expression changes in fatty acid metabolism and the observed expression changes of genes involved in transsulfuration pathway and methionine cycle are linked.

Increased transcripts of estrogen receptor 1 and vitellogenin 5 were observed in male progeny livers associated with parental high dietary ARA levels. Interestingly, the same transcripts showed oppositely directed gene expression (regardless the chosen threshold) in the parental fish (F0 high ARA) suggesting compensatory mechanisms being involved. Vitellogenin, a lipid transporting protein, is produced in the liver and transported to the ovary as an egg yolk protein, under the regulation of estrogens in female fish [7779]. Dietary ARA has been suggested to affect oocyte maturation and especially steroidogenesis in fish, but knowledge on mechanisms and impact on the following generations is lacking [8082]. Induction of vitellogenin in male fish is commonly used as a marker for endocrine disruption [28] and studies on zebrafish have shown that vitellogenin synthesis can be induced in male fish when exposed to an estrogen (steroid hormone) named 17β-estradiol [83, 84]. In addition, vitellogenin has also an immunological role facilitating the defence against virus and bacteria [85, 86]. The exact biological role of vitellogenin in male fish is not clear. In humans, actions of estrogens are mediated by estrogen receptors and their role has been linked to metabolic inflammation [87]. Estrogens can regulate various energy metabolism pathways and disturbance by endocrine disruptors has been discussed in development of obesity [88]. However, our results from livers of male zebrafish progeny suggest that hepatic gene regulation related to steroidogenesis and estrogen signalling are sensitive to parental dietary high ARA intake.

We hypothesise that the observed DEGs in F1 high ARA livers compared to the control group, were associated with parental diet, but the exact modulatory mechanisms are not known. One mechanism whereby maternal diet can impact the progeny’s physiological status is the contribution to the nutrient reservoir in the developing oocyte [28, 89, 90]. Maternal transcripts present in the fertilized egg can steer gene expression in the developing embryo [31, 91, 92]. Epigenetic modifications of the genome in the gametes have also been identified as likely mechanisms through which environmental conditions, such as diet, can affect progeny transcriptomic patterns [26, 93]. Previous studies on zebrafish have shown that nutritional status of the parents such as vitamin B or vitamin E deficiency altered the transcriptome of their embryos [32, 33] and the livers of adult offspring [34]. Dietary PUFAs have been shown to affect membrane composition, cell signalling, gene expression and metabolism of the developing oocyte and thereby being able to influence the development of the next generation [7, 38, 94, 95].

Zebrafish has become a favoured research tool to investigate both molecular biological processes and the importance of nutrition in developmental, health and disease outcomes in vertebrates [46, 51, 96, 97]. Due to genetic, anatomical and physiological similarities to other vertebrates, zebrafish can be a useful model to evaluate the influence of dietary profiles on gene expression that can persist throughout life and across multiple generations as shown for different teleost species before [32, 33, 47, 52, 98]. Here, we demonstrated that parental diet affects the hepatic transcriptomic profiles in adult progeny. At the transcriptional level, we found that parental high ARA had a greater effect in the progeny than in the parents who were directly exposed to the experimental diets. These results are surprising, and show that the nutritional priming from parental generation has substantial effect on the progeny transcriptional profile. However, other factors could have influenced the lack of differential expression in the parental generation such as previous dietary treatments and time of sampling in relation to feeding. More knowledge is needed on how parental dietary habits can shape the progeny’s transcriptome and thereby possibly alter metabolic pathways in the progeny. Whether these differences can be inherited to further generations is an area for further research.


Our work has shown that the parental diet modulated the transcription of a range of genes in the adult progeny connected to the fatty acid and retinoid metabolism, methionine cycle, transsulfuration pathway and estrogen signalling. We cannot distinguish if the effect at the transcriptional level is due to the nutritional composition of eggs (oocytes), maternal mRNA deposition or progeny transcriptome potential through programming of the gametes. Modulation of the transcriptome at early life stages can in turn affect metabolic pathways and their activity at later life stages. To our knowledge, the present study is the first one investigating liver transcriptome characteristics of adult zebrafish progeny from parents fed high ARA levels. Nevertheless, further study is required to understand deeper mechanisms on how those gene expression differences observed in adult progeny develop and if these effects can be transferred to future generations.

Supporting information

S1 File. Composition of control and high ARA diet.


S2 File. Danio rerio primer sequences used for real-time RT-qPCR of target genes for RNA-sequencing validation.


S3 File. Gene counts and overlap of significant differentially expressed genes (adjusted p<0.05) in F0 and F1 zebrafish livers obtained from RNA-sequencing and read mapping to the RefSeq and Ensembl reference genome (GRCz10).


S4 File.

Volcano plot of RNA-sequencing data from male livers comparing control (A) and high ARA (B) group in F0 and F1 generation.


S5 File. Ingenuity® pathway analysis identified lipid metabolism associated biological functions represented by differentially expressed genes (adjusted p<0.05) from RNA-sequencing of F1 high ARA compared to control livers.


S1 Table. Differentially expressed genes in male F0 high ARA livers compared to control livers using the RefSeq genome.


S2 Table. Differentially expressed genes in male F0 high ARA livers compared to control livers using the Ensembl genome.


S3 Table. Differentially expressed genes in male F1 high ARA livers compared to control livers using the RefSeq genome.


S4 Table. Differentially expressed genes in male F1 high ARA livers compared to control livers using the Ensembl genome.


S5 Table. Differentially expressed genes in male F0 control livers compared to F1 control livers using the RefSeq genome.


S6 Table. Differentially expressed genes in male F0 control livers compared to F1 control livers using the Ensembl genome.


S7 Table. Differentially expressed genes in male F0 high ARA livers compared to F1 high ARA livers using the RefSeq genome.


S8 Table. Differentially expressed genes in male F0 high ARA livers compared to F1 high ARA livers using the Ensembl genome.


S9 Table. Concordant genes from the comparison of differentially expressed genes (adjusted p<0.05) between F0 control and F0 high ARA livers after mapping to the RefSeq and Ensembl reference genome (GRCz10).


S10 Table. Concordant genes from the comparison of differentially expressed genes (adjusted p<0.05) between F1 control and F1 high ARA livers after mapping to the RefSeq and Ensembl reference genome (GRCz10).


S11 Table. Concordant genes from the comparison of differentially expressed genes (adjusted p<0.05) between F0 and F1 control livers after mapping to the RefSeq and Ensembl reference genome (GRCz10).


S12 Table. Concordant genes from the comparison of differentially expressed genes (adjusted p<0.05) between F0 and F1 high ARA livers after mapping to the RefSeq and Ensembl reference genome (GRCz10).


S13 Table. Concordant genes in F1 generation uploaded into the Ingenuity® Pathway Analysis software suite using human orthologues.


S14 Table. Functional annotation of annotated differentially expressed genes between both F0 and F1 control and between F0 and F1 high ARA groups for KEGG pathways and GO terms.



We are grateful to Øyvind Reinshol and Synnøve Wintertun for assistance at the in-house zebrafish facilities and Eva Mykkeltvedt for RNA isolation at NIFES. Library preparation and sequencing was performed by the Norwegian High-Throughput Sequencing Centre in Oslo (NSC), a national technology platform hosted by the University of Oslo.


  1. 1. Sanders TA. Polyunsaturated fatty acids in the food chain in Europe. Am J Clin Nutr. 2000 Jan;71(1 Suppl):176S–8S.
  2. 2. Simopoulos AP. Evolutionary aspects of diet, the omega-6/omega-3 ratio and genetic variation: nutritional implications for chronic diseases. Biomed Pharmacother. 2006 Nov;60(9):502–7. pmid:17045449
  3. 3. Patterson E, Wall R, Fitzgerald GF, Ross RP, Stanton C. Health implications of high dietary omega-6 polyunsaturated Fatty acids. J Nutr Metab. 2012;2012:539426. pmid:22570770
  4. 4. Candela CG, Lopez LMB, Kohen VL. Importance of a balanced omega 6/omega 3 ratio for the maintenance of health. Nutritional recommendations. Nutr Hosp. 2011 Mar-Apr;26(2):323–9. pmid:21666970
  5. 5. Simopoulos AP. The importance of the omega-6/omega-3 fatty acid ratio in cardiovascular disease and other chronic diseases. Exp Biol Med (Maywood). 2008 Jun;233(6):674–88.
  6. 6. Simopoulos AP. An Increase in the Omega-6/Omega-3 Fatty Acid Ratio Increases the Risk for Obesity. Nutrients. 2016 Mar;8(3):128. pmid:26950145
  7. 7. Massiera F, Barbry P, Guesnet P, Joly A, Luquet S, Moreilhon-Brest C, et al. A Western-like fat diet is sufficient to induce a gradual enhancement in fat mass over generations. J Lipid Res. 2010 Aug;51(8):2352–61. pmid:20410018
  8. 8. Koletzko B, Larque E, Demmelmair H. Placental transfer of long-chain polyunsaturated fatty acids (LC-PUFA). J Perinat Med. 2007;35 Suppl 1:S5–11.
  9. 9. Wakefield SL, Lane M, Schulz SJ, Hebart ML, Thompson JG, Mitchell M. Maternal supply of omega-3 polyunsaturated fatty acids alter mechanisms involved in oocyte and early embryo development in the mouse. Am J Physiol Endocrinol Metab. 2008 Feb;294(2):E425–34. pmid:18073322
  10. 10. Lie KK, Kvalheim K, Rasinger JD, Harboe T, Nordgreen A, Moren M. Vitamin A and arachidonic acid altered the skeletal mineralization in Atlantic cod (Gadus morhua) larvae without any interactions on the transcriptional level. Comp Biochem Phys A. 2016 Jan;191:80–8.
  11. 11. de Vrieze E, Moren M, Metz JR, Flik G, Lie KK. Arachidonic acid enhances turnover of the dermal skeleton: studies on zebrafish scales. PLoS One. 2014;9(2):e89347. pmid:24586706
  12. 12. Buczynski MW, Dumlao DS, Dennis EA. Thematic Review Series: Proteomics. An integrated omics analysis of eicosanoid biology. J Lipid Res. 2009 Jun;50(6):1015–38. pmid:19244215
  13. 13. Brash AR. Arachidonic acid as a bioactive molecule. J Clin Invest. 2001 Jun;107(11):1339–45. pmid:11390413
  14. 14. Harizi H, Corcuff JB, Gualde N. Arachidonic-acid-derived eicosanoids: roles in biology and immunopathology. Trends Mol Med. 2008 Oct;14(10):461–9. pmid:18774339
  15. 15. Kliewer SA, Sundseth SS, Jones SA, Brown PJ, Wisely GB, Koble CS, et al. Fatty acids and eicosanoids regulate gene expression through direct interactions with peroxisome proliferator-activated receptors alpha and gamma. Proc Natl Acad Sci U S A. 1997 Apr 29;94(9):4318–23. pmid:9113987
  16. 16. Schmitz G, Ecker J. The opposing effects of n-3 and n-6 fatty acids. Prog Lipid Res. 2008 Mar;47(2):147–55. pmid:18198131
  17. 17. Jump DB. Fatty acid regulation of gene transcription. Crit Rev Clin Lab Sci. 2004;41(1):41–78. pmid:15077723
  18. 18. Silva-Martinez GA, Rodriguez-Rios D, Alvarado-Caudillo Y, Vaquero A, Esteller M, Carmona FJ, et al. Arachidonic and oleic acid exert distinct effects on the DNA methylome. Epigenetics. 2016 May 3;11(5):321–34. pmid:27088456
  19. 19. Benatti P, Peluso G, Nicolai R, Calvani M. Polyunsaturated fatty acids: biochemical, nutritional and epigenetic properties. J Am Coll Nutr. 2004 Aug;23(4):281–302. pmid:15310732
  20. 20. Niculescu MD, Lupu DS, Craciunescu CN. Perinatal manipulation of alpha-linolenic acid intake induces epigenetic changes in maternal and offspring livers. FASEB J. 2013 Jan;27(1):350–8. pmid:22997227
  21. 21. Adam AC, Lie KK, Moren M, Skjaerven KH. High dietary arachidonic acid levels induce changes in complex lipids and immune-related eicosanoids and increase levels of oxidised metabolites in zebrafish (Danio rerio). Br J Nutr. 2017 May 09:1–11.
  22. 22. Rando OJ, Simmons RA. I'm eating for two: parental dietary effects on offspring metabolism. Cell. 2015 Mar 26;161(1):93–105. pmid:25815988
  23. 23. Hanley B, Dijane J, Fewtrell M, Grynberg A, Hummel S, Junien C, et al. Metabolic imprinting, programming and epigenetics—a review of present priorities and future opportunities. Br J Nutr. 2010 Jul;104 Suppl 1:S1–25.
  24. 24. Lane M, Zander-Fox DL, Robker RL, McPherson NO. Peri-conception parental obesity, reproductive health, and transgenerational impacts. Trends Endocrinol Metab. 2015 Feb;26(2):84–90. pmid:25523615
  25. 25. Watkins AJ, Lucas ES, Fleming TP. Impact of the periconceptional environment on the programming of adult disease. J Dev Orig Health Dis. 2010 Apr;1(2):87–95. pmid:25143062
  26. 26. Lillycrop KA. Effect of maternal diet on the epigenome: implications for human metabolic disease. Proc Nutr Soc. 2011 Feb;70(1):64–72. pmid:21266093
  27. 27. Gu L, Liu H, Gu X, Boots C, Moley KH, Wang Q. Metabolic control of oocyte development: linking maternal nutrition and reproductive outcomes. Cell Mol Life Sci. 2015 Jan;72(2):251–71. pmid:25280482
  28. 28. Arukwe A, Goksoyr A. Eggshell and egg yolk proteins in fish: hepatic proteins for the next generation: oogenetic, population, and evolutionary implications of endocrine disruption. Comp Hepatol. 2003 Mar 06;2(1):4. pmid:12685931
  29. 29. Tadros W, Lipshitz HD. The maternal-to-zygotic transition: a play in two acts. Development. 2009 Sep;136(18):3033–42. pmid:19700615
  30. 30. Aanes H, Winata CL, Lin CH, Chen JQP, Srinivasan KG, Lee SGP, et al. Zebrafish mRNA sequencing deciphers novelties in transcriptome dynamics during maternal to zygotic transition. Genome Research. 2011 Aug;21(8):1328–38. pmid:21555364
  31. 31. Liebers R, Rassoulzadegan M, Lyko F. Epigenetic regulation by heritable RNA. PLoS Genet. 2014 Apr;10(4):e1004296. pmid:24743450
  32. 32. Skjaerven KH, Jakt LM, Dahl JA, Espe M, Aanes H, Hamre K, et al. Parental vitamin deficiency affects the embryonic gene expression of immune-, lipid transport- and apolipoprotein genes. Sci Rep. 2016 Oct 12;6:34535. pmid:27731423
  33. 33. Miller GW, Truong L, Barton CL, Labut EM, Lebold KM, Traber MG, et al. The influences of parental diet and vitamin E intake on the embryonic zebrafish transcriptome. Comp Biochem Physiol Part D Genomics Proteomics. 2014 Jun;10:22–9. pmid:24657723
  34. 34. Skjaerven KH, Jakt LM, Fernandes JMO, Dahl JA, Adam AC, Klughammer J, et al. Parental micronutrient deficiency distorts liver DNA methylation and expression of lipid genes associated with a fatty-liver-like phenotype in offspring. Sci Rep. 2018 Feb 14;8(1):3055. pmid:29445184
  35. 35. Newman T, Jhinku N, Meier M, Horsfield J. Dietary Intake Influences Adult Fertility and Offspring Fitness in Zebrafish. PLoS One. 2016;11(11):e0166394. pmid:27870856
  36. 36. Warzych E, Cieslak A, Pawlak P, Renska N, Pers-Kamczyc E, Lechniak D. Maternal nutrition affects the composition of follicular fluid and transcript content in gilt oocytes. Vet Med-Czech. 2011;56(4):156–67.
  37. 37. Wonnacott KE, Kwong WY, Hughes J, Salter AM, Lea RG, Garnsworthy PC, et al. Dietary omega-3 and -6 polyunsaturated fatty acids affect the composition and development of sheep granulosa cells, oocytes and embryos. Reproduction. 2010 Jan;139(1):57–69. pmid:19789173
  38. 38. McMillen IC, MacLaughlin SM, Muhlhausler BS, Gentili S, Duffield JL, Morrison JL. Developmental origins of adult health and disease: the role of periconceptional and foetal nutrition. Basic Clin Pharmacol Toxicol. 2008 Feb;102(2):82–9. pmid:18226059
  39. 39. O'Callaghan D, Yaakub H, Hyttel P, Spicer LJ, Boland MP. Effect of nutrition and superovulation on oocyte morphology, follicular fluid composition and systemic hormone concentrations in ewes. J Reprod Fertil. 2000 Mar;118(2):303–13. pmid:10864794
  40. 40. Sorbera LA, Asturiano JF, Carrillo M, Zanuy S. Effects of polyunsaturated fatty acids and prostaglandins on oocyte maturation in a marine teleost, the European sea bass (Dicentrarchus labrax). Biol Reprod. 2001 Jan;64(1):382–9. pmid:11133697
  41. 41. Bruce M, Oyen F, Bell G, Asturiano JF, Farndale B, Carrillo M, et al. Development of broodstock diets for the European Sea Bass (Dicentrarchus labrax) with special emphasis on the importance of n-3 and n-6 highly unsaturated fatty acid to reproductive performance. Aquaculture. 1999 Jul 1;177(1–4):85–97.
  42. 42. Asil SM, Kenari AA, Miyanji GR, Van Der Kraak G. The influence of dietary arachidonic acid on growth, reproductive performance, and fatty acid composition of ovary, egg and larvae in an anabantid model fish, Blue gourami (Trichopodus trichopterus; Pallas, 1770). Aquaculture. 2017 Jul 1;476:8–18.
  43. 43. Furuita H, Yamamoto T, Shima T, Suzuki N, Takeuchi T. Effect of arachidonic acid levels in broodstock diet on larval and egg quality of Japanese flounder Paralichthys olivaceus. Aquaculture. 2003 Apr 14;220(1–4):725–35.
  44. 44. Jaya-Ram A, Kuah MK, Lim PS, Kolkovski S, Shu-Chien AC. Influence of dietary HUFA levels on reproductive performance, tissue fatty acid profile and desaturase and elongase mRNAs expression in female zebrafish Danio rerio. Aquaculture. 2008 Jun 3;277(3–4):275–81.
  45. 45. Anderson JL, Carten JD, Farber SA. Zebrafish lipid metabolism: from mediating early patterning to the metabolism of dietary fat and cholesterol. Methods Cell Biol. 2011;101:111–41. pmid:21550441
  46. 46. Schlegel A, Stainier DY. Lessons from "lower" organisms: what worms, flies, and zebrafish can teach us about human energy metabolism. PLoS Genet. 2007 Nov;3(11):e199. pmid:18081423
  47. 47. Howe K, Clark MD, Torroja CF, Torrance J, Berthelot C, Muffato M, et al. The zebrafish reference genome sequence and its relationship to the human genome. Nature. 2013 Apr 25;496(7446):498–503. pmid:23594743
  48. 48. Dahm R, Geisler R. Learning from small fry: the zebrafish as a genetic model organism for aquaculture fish species. Mar Biotechnol (NY). 2006 Jul-Aug;8(4):329–45.
  49. 49. Alestrom P, Holter JL, Nourizadeh-Lillabadi R. Zebrafish in functional genomics and aquatic biomedicine. Trends Biotechnol. 2006 Jan;24(1):15–21. pmid:16309768
  50. 50. Baker TR, Peterson RE, Heideman W. Using zebrafish as a model system for studying the transgenerational effects of dioxin. Toxicol Sci. 2014 Apr;138(2):403–11. pmid:24470537
  51. 51. Watts SA, Lawrence C, Powell M, D'Abramo LR. The Vital Relationship Between Nutrition and Health in Zebrafish. Zebrafish. 2016 Jul;13 Suppl 1:S72–6.
  52. 52. Ulloa PE, Medrano JF, Feijoo CG. Zebrafish as animal model for aquaculture nutrition research. Front Genet. 2014;5:313. pmid:25309575
  53. 53. Williams CR, Baccarella A, Parrish JZ, Kim CC. Trimming of sequence reads alters RNA-Seq gene expression estimates. BMC Bioinformatics. 2016 Feb 25;17:103. pmid:26911985
  54. 54. Yates A, Akanni W, Amode MR, Barrell D, Billis K, Carvalho-Silva D, et al. Ensembl 2016. Nucleic Acids Res. 2016 Jan 04;44(D1):D710–6. pmid:26687719
  55. 55. Kim D, Langmead B, Salzberg SL. HISAT: a fast spliced aligner with low memory requirements. Nat Methods. 2015 Apr;12(4):357–60. pmid:25751142
  56. 56. Liao Y, Smyth GK, Shi W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics. 2014 Apr 01;30(7):923–30. pmid:24227677
  57. 57. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15(12):550. pmid:25516281
  58. 58. Zhao S, Zhang B. A comprehensive evaluation of ensembl, RefSeq, and UCSC annotations in the context of RNA-seq read mapping and gene quantification. BMC Genomics. 2015 Feb 18;16:97. pmid:25765860
  59. 59. Edgar R, Domrachev M, Lash AE. Gene Expression Omnibus: NCBI gene expression and hybridization array data repository. Nucleic Acids Res. 2002 Jan 01;30(1):207–10. pmid:11752295
  60. 60. Yu G, Wang LG, Han Y, He QY. clusterProfiler: an R package for comparing biological themes among gene clusters. OMICS. 2012 May;16(5):284–7. pmid:22455463
  61. 61. Skjaerven KH, Olsvik PA, Finn RN, Holen E, Hamre K. Ontogenetic expression of maternal and zygotic genes in Atlantic cod embryos under ambient and thermally stressed conditions. Comp Biochem Physiol A Mol Integr Physiol. 2011 Jun;159(2):196–205. pmid:21377533
  62. 62. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A, et al. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol. 2002 Jun 18;3(7):RESEARCH0034.
  63. 63. Olsvik PA, Williams TD, Tung HS, Mirbahai L, Sanden M, Skjaerven KH, et al. Impacts of TCDD and MeHg on DNA methylation in zebrafish (Danio rerio) across two generations. Comp Biochem Physiol C Toxicol Pharmacol. 2014 Sep;165:17–27. pmid:24878852
  64. 64. Raser JM O'Shea EK. Noise in gene expression: origins, consequences, and control. Science. 2005 Sep 23;309(5743):2010–3. pmid:16179466
  65. 65. Cunningham TJ, Duester G. Mechanisms of retinoic acid signalling and its roles in organ and limb development. Nat Rev Mol Cell Bio. 2015 Feb;16(2):110–23.
  66. 66. Raverdeau M, Gely-Pernot A, Feret B, Dennefeld C, Benoit G, Davidson I, et al. Retinoic acid induces Sertoli cell paracrine signals for spermatogonia differentiation but cell autonomously drives spermatocyte meiosis. P Natl Acad Sci USA. 2012 Oct 9;109(41):16582–7.
  67. 67. Vandersea MW, Fleming P, McCarthy RA, Smith DG. Fin duplications and deletions induced by disruption of retinoic acid signaling. Development Genes and Evolution. 1998 Apr;208(2):61–8. pmid:9569347
  68. 68. Rhinn M, Dolle P. Retinoic acid signalling during development. Development. 2012 Mar 1;139(5):843–58. pmid:22318625
  69. 69. Lee LM, Leung CY, Tang WW, Choi HL, Leung YC, McCaffery PJ, et al. A paradoxical teratogenic mechanism for retinoic acid. Proc Natl Acad Sci U S A. 2012 Aug 21;109(34):13668–73. pmid:22869719
  70. 70. Conaway HH, Henning P, Lerner UH. Vitamin a metabolism, action, and role in skeletal homeostasis. Endocr Rev. 2013 Dec;34(6):766–97. pmid:23720297
  71. 71. Kota BP, Huang THW, Roufogalis BD. An overview on biological mechanisms of PPARs. Pharmacological Research. 2005 Feb;51(2):85–94. pmid:15629253
  72. 72. Ziouzenkova O, Plutzky J. Retinoid metabolism and nuclear receptor responses: New insights into coordinated regulation of the PPAR-RXR complex. FEBS Lett. 2008 Jan 09;582(1):32–8. pmid:18068127
  73. 73. Rakhshandehroo M, Knoch B, Muller M, Kersten S. Peroxisome proliferator-activated receptor alpha target genes. PPAR Res. 2010;2010.
  74. 74. da Silva RP, Kelly KB, Al Rajabi A, Jacobs RL. Novel insights on interactions between folate and lipid metabolism. Biofactors. 2014 May-Jun;40(3):277–83. pmid:24353111
  75. 75. Lysne V, Strand E, Svingen GF, Bjorndal B, Pedersen ER, Midttun O, et al. Peroxisome Proliferator-Activated Receptor Activation is Associated with Altered Plasma One-Carbon Metabolites and B-Vitamin Status in Rats. Nutrients. 2016 Jan 05;8(1).
  76. 76. Huang T, Hu X, Khan N, Yang J, Li D. Effect of polyunsaturated fatty acids on homocysteine metabolism through regulating the gene expressions involved in methionine metabolism. ScientificWorldJournal. 2013;2013:931626. pmid:23766724
  77. 77. Nagler JJ, Davis TL, Modi N, Vijayan MM, Schultz I. Intracellular, not membrane, estrogen receptors control vitellogenin synthesis in the rainbow trout. Gen Comp Endocrinol. 2010 Jun 01;167(2):326–30. pmid:20346361
  78. 78. Hara A, Hiramatsu N, Fujita T. Vitellogenesis and choriogenesis in fishes. Fisheries Sci. 2016 Mar;82(2):187–202.
  79. 79. Avarre JC, Lubzens E, Babin PJ. Apolipocrustacein, formerly vitellogenin, is the major egg yolk precursor protein in decapod crustaceans and is homologous to insect apolipophorin II/I and vertebrate apolipoprotein B. BMC Evol Biol. 2007 Jan 22;7:3. pmid:17241455
  80. 80. Xu HG, Cao L, Zhang YQ, Johnson RB, Wei YL, Zheng KK, et al. Dietary arachidonic acid differentially regulates the gonadal steroidogenesis in the marine teleost, tongue sole (Cynoglossus semilaevis), depending on fish gender and maturation stage. Aquaculture. 2017 Feb 1;468:378–85.
  81. 81. Van der Kraak G, Chang JP. Arachidonic acid stimulates steroidogenesis in goldfish preovulatory ovarian follicles. Gen Comp Endocrinol. 1990 Feb;77(2):221–8. pmid:2106468
  82. 82. Norberg B, Kleppe L, Andersson E, Thorsen A, Rosenlund G, Hamre K. Effects of dietary arachidonic acid on the reproductive physiology of female Atlantic cod (Gadus morhua L.). Gen Comp Endocrinol. 2017 May 30;250:21–35. pmid:28576420
  83. 83. Rose J, Holbech H, Lindholst C, Norum U, Povlsen A, Korsgaard B, et al. Vitellogenin induction by 17beta-estradiol and 17alpha-ethinylestradiol in male zebrafish (Danio rerio). Comp Biochem Physiol C Toxicol Pharmacol. 2002 Apr;131(4):531–9. pmid:11976068
  84. 84. Brion F, Tyler CR, Palazzi X, Laillet B, Porcher JM, Garric J, et al. Impacts of 17beta-estradiol, including environmentally relevant concentrations, on reproduction after exposure during embryo-larval-, juvenile- and adult-life stages in zebrafish (Danio rerio). Aquat Toxicol. 2004 Jun 24;68(3):193–217. pmid:15159048
  85. 85. Sun C, Zhang S. Immune-Relevant and Antioxidant Activities of Vitellogenin and Yolk Proteins in Fish. Nutrients. 2015 Oct 22;7(10):8818–29. pmid:26506386
  86. 86. Zhang S, Wang S, Li H, Li L. Vitellogenin, a multivalent sensor and an antimicrobial effector. Int J Biochem Cell Biol. 2011 Mar;43(3):303–5. pmid:21075213
  87. 87. Monteiro R, Teixeira D, Calhau C. Estrogen Signaling in Metabolic Inflammation. Mediat Inflamm. 2014.
  88. 88. Chen JQ, Brown TR, Russo J. Regulation of energy metabolism pathways by estrogens and estrogenic chemicals and potential implications in obesity associated with increased exposure to endocrine disruptors. Bba-Mol Cell Res. 2009 Jul;1793(7):1128–43.
  89. 89. Leroy JLMR, Valckx S, Sturmey R, Bossaert P, Van Hoeck V, Bols PEJ. Maternal metabolic health and oocyte quality: the role of the intrafollicular environment. Anim Reprod. 2012 Oct-Dec;9(4):777–88.
  90. 90. Hamre K, Yufera M, Ronnestad I, Boglione C, Conceicao LEC, Izquierdo M. Fish larval nutrition and feed formulation: knowledge gaps and bottlenecks for advances in larval rearing. Rev Aquacult. 2013 May;5:S26–S58.
  91. 91. Schier AF. The maternal-zygotic transition: Death and birth of RNAs. Science. 2007 Apr 20;316(5823):406–7. pmid:17446392
  92. 92. Harvey SA, Sealy I, Kettleborough R, Fenyes F, White R, Stemple D, et al. Identification of the zebrafish maternal and paternal transcriptomes. Development. 2013 Jul;140(13):2703–10. pmid:23720042
  93. 93. Fowden AL, Coan PM, Angiolini E, Burton GJ, Constancia M. Imprinted genes and the epigenetic regulation of placental phenotype. Prog Biophys Mol Biol. 2011 Jul;106(1):281–8. pmid:21108957
  94. 94. McKeegan PJ, Sturmey RG. The role of fatty acids in oocyte and early embryo development. Reprod Fertil Dev. 2011;24(1):59–67. pmid:22394718
  95. 95. Muhlhausler BS, Ailhaud GP. Omega-6 polyunsaturated fatty acids and the early origins of obesity. Curr Opin Endocrinol Diabetes Obes. 2013 Feb;20(1):56–61. pmid:23249760
  96. 96. Ribas L, Piferrer F. The zebrafish (Danio rerio) as a model organism, with emphasis on applications for finfish aquaculture research. Rev Aquacult. 2014 Dec;6(4):209–40.
  97. 97. Grunwald DJ, Eisen JS. Headwaters of the zebrafish—emergence of a new model vertebrate. Nat Rev Genet. 2002 Sep;3(9):717–24. pmid:12209146
  98. 98. Seiliez I, Velez EJ, Lutfi E, Dias K, Plagnes-Juan E, Marandel L, et al. Eating for two: Consequences of parental methionine nutrition on offspring metabolism in rainbow trout (Oncorhynchus mykiss). Aquaculture. 2017 Mar 20;471:80–91.