Gene 543 (2014) 268–274

Contents lists available at ScienceDirect

Gene journal homepage: www.elsevier.com/locate/gene

The evolution and functional divergence of the beta-carotene oxygenase gene family in teleost fish—Exemplified by Atlantic salmon Hanna Helgeland a,⁎, Simen Rød Sandve b, Jacob Seilø Torgersen c, Mari Kyllesø Halle a, Hilde Sundvold a, Stig Omholt a, Dag Inge Våge a a b c

Centre for Integrative Genetics (CIGENE), Dept. of Animal and Aquacultural Sciences, Norwegian University of Life Sciences (UMB), P.O. Box 5003, N-1432 As, Norway Centre for Integrative Genetics (CIGENE), Dept. of Plant and Environmental Sciences, Norwegian University of Life Sciences (UMB), P.O. Box 5003, N-1432 As, Norway Nofima AS, PO Box 5010, N-1432 Ås, Norway

a r t i c l e

i n f o

Article history: Received 30 September 2013 Received in revised form 14 February 2014 Accepted 21 February 2014 Available online 26 February 2014 Keywords: Beta-carotene oxygenase Gene family Genome duplication Teleost fish Salmon

a b s t r a c t In mammals, two carotenoid cleaving oxygenases are known; beta-carotene 15,15′-monooxygenase (BCMO1) and beta-carotene 9′,10′-oxygenase (BCO2). BCMO1 is a key enzyme in vitamin A synthesis by symmetrically cleaving beta-carotene into 2 molecules of all-trans-retinal, while BCO2 is responsible for asymmetric cleavage of a broader range of carotenoids. Here, we show that the Atlantic salmon beta-carotene oxygenase (bco) gene family contains 5 members, three bco2 and two bcmo1 paralogs. Using public sequence databases, multiple bco genes were also found in several additional teleost species. Phylogenetic analysis indicates that bco2a and bco2b originate from the teleost fish specific genome duplication (FSGD or 3R), while the third and more distant paralog, bco2 like, might stem from a prior duplication event in the teleost lineage. The two bcmo1 paralogs (bcmo1 and bcmo1 like) appear to be the result of an ancient duplication event that took place before the divergence of ray-finned (Actinopterygii) and lobe-finned fish (Sarcopterygii), with subsequent nonfunctionalization and loss of one Sarcopterygii paralog. Gene expression analysis of the bcmo1 and bco2 paralogs in Atlantic salmon reveals regulatory divergence with tissue specific expression profiles, suggesting that the beta-carotene oxygenase subtypes have evolved functional divergences. We suggest that teleost fish have evolved and maintained an extended repertoire of beta-carotene oxygenases compared to the investigated Sarcopterygii species, and hypothesize that the main driver behind this functional divergence is the exposure to a diverse set of carotenoids in the aquatic environment. © 2014 Elsevier B.V. All rights reserved.

1. Introduction The mammalian family of carotenoid cleaving oxygenases contains two members; beta-carotene 15,15′-monooxygenase (BCMO1) and beta-carotene 9′,10′-oxygenase (BCO2) (Lobo et al., 2011). BCMO1 serves a key function in vitamin A synthesis by cleaving beta-carotene, at the 15,15′ position, into 2 molecules of retinal (von Lintig and Vogt,

Abbreviations: aa, amino acid(s); ANOVA, analysis of variance; BCMO1, beta-carotene 15,15′-monooxygenase; bco, beta-carotene oxygenase; BCO2, beta-carotene 9′,10′-oxygenase; bp, base pair(s); cDNA, DNA complementary to RNA; ef1a, elongation factor 1alpha; EST, expressed sequence tag; FSGD, fish-specific genome duplication; ICSASG, International Collaboration to Sequence the Atlantic Salmon Genome; MAFFT, Multiple Alignment using Fast Fourier Transform; MEGA, Molecular Evolutionary Genetics Analysis; ML, maximum likelihood; MYA, million years ago; NCBI, National Center for Biotechnology Information; NIVA, Norwegian Institute for Water Research; PCR, polymerase chain reaction; qRT-PCR, quantitative real-time PCR; RPE65, retinal pigment epithelium-specific 65 kDa protein. ⁎ Corresponding author.

http://dx.doi.org/10.1016/j.gene.2014.02.042 0378-1119/© 2014 Elsevier B.V. All rights reserved.

2000; Wyss et al., 2000). The substrates of BCMO1 are pro-vitamin A carotenoids containing at least one unsubstituted beta-ionone ring, such as beta-carotene and beta-cryptoxanthin (Lindqvist and Andersson, 2002). BCO2, also named BCDO2, is responsible for asymmetric cleavage at the 9′–10′ double bond of the polyene backbone of its substrates (Kiefer et al., 2001), and is suggested as an alternative pathway to retinoic acid synthesis (Simões-costa et al., 2008). BCO2 displays broader substrate specificity than BCMO1, and is shown to provide oxidative cleavage of both carotenes, including lycopene, and xanthophylls like zeaxanthin, lutein and beta-cryptoxanthin in mammals (Kiefer et al., 2001; Kim et al., 2011; Mein et al., 2011). In contrast to BCMO1 which is a cytoplasmic protein (Lindqvist and Andersson, 2002), BCO2 was recently found to be expressed in the mitochondria where it acts as a carotenoid scavenger, providing protection from carotenoidinduced mitochondrial dysfunction (Amengual et al., 2011; Lobo et al., 2012). Two genome duplications are hypothesized to have occurred early in the vertebrate lineage (the 2R hypothesis). Analysis of the sea lamprey genome and comparisons to genomes of jawed vertebrates

H. Helgeland et al. / Gene 543 (2014) 268–274

indicates that the two genome duplications likely took place in a common ancestor, before the divergence of jawed vertebrates (Smith et al., 2013). Several studies provide evidence for a fish-specific genome duplication (FSGD or 3R) estimated to have occurred approximately 350 Ma ago (Christoffels et al., 2004; Jaillon et al., 2004). Finally a whole genome duplication event occurred in the salmonid lineage 25–100 MYA, rendering salmonids partially tetraploid (Allendorf & Thorgaard, 1984; Lien et al., 2011). In the present study, we have identified and characterized five beta-carotene oxygenase (bco) genes in Atlantic salmon. In addition, we have used public genome databases to identify other piscine betacarotene oxygenases. The piscine bco sequences were used to investigate the evolutionary relationships within the teleost family of bco genes and in relation to Sarcopterygii orthologs. 2. Material and methods 2.1. Animals Six healthy Atlantic salmon with an average weight of 1250 g were sampled from a research fish farm run by the Norwegian Institute for Water Research (NIVA). The fish were fed ad libitum commercial feed (Skretting), containing 50 mg/kg astaxanthin. The muscle, liver and intestinal tissue from the midgut were flash-frozen in liquid nitrogen and stored at −80 °C until RNA extraction. 2.2. RNA isolation and cDNA synthesis Total RNA was isolated from salmon muscle tissue using TRIzol Reagent (Invitrogen) and purified with the RNeasy soft tissue Mini Kit (Qiagen). RNA was treated with Invitrogen DNase and RNaseOUT™

269

Recombinant Ribonuclease Inhibitor, protocol according to the manufacturer. Sufficient quantity and quality of the total RNA were confirmed using NanoDrop (Thermo Scientific) and Bioanalyzer (Agilent Technologies). cDNA was synthesized using SuperScript® III Reverse Transcriptase (Invitrogen), a poly dT-primer for sequencing or a 1:1 mix of random hexamers and poly dT for qRT-PCR, and 500 ng RNA, protocol according to the manufacturer. 2.3. In silico bco sequence mining Zebrafish bco reference sequences from the NCBI database (http:// www.ncbi.nlm.nih.gov/), NM_001040312 and NM_131798, were queried using BLASTN against the NCBI BLAST EST database, The Gene Index Project and salmon trace archives, a de novo draft salmon genome based on sequences provided by the International Collaboration to Sequence the Atlantic Salmon Genome (ICSASG) (Davidson et al., 2010), to identify salmon bco sequences. Obtained salmonid bco sequences were continuously included as query to extend the search. Bcmo1 and bco2 sequences from additional species were identified in the NCBI, UCSC and Ensembl databases. Chromosomal positions of the medaka, Tetraodon and zebrafish bco2 and bcmo1 orthologs were obtained from the UCSC Genome Bioinformatics Site. 2.4. Amplification and sequencing of Atlantic salmon bco2 and bcmo1 genes Primers were designed in Primer3Plus (Untergasser et al., 2007) based on sequences obtained from trace archives and the following ESTs: TC184762, BT026919.1, EV366861.1, CA041600, CB516394, and TC109507. A cDNA sequence for the bco2 like gene was kindly provided by Dr. Ben Koop (University of Victoria, Canada). The coding region of the salmon bco2a gene was amplified with primers P1 + P2 and P3 +

Fig. 1. Phylogenetic tree of Bcmo1 proteins in ray-finned and lobe-finned species, based on amino acid alignment. Amphioxus (Branchiostoma floridae) is included as outgroup.

270

H. Helgeland et al. / Gene 543 (2014) 268–274

P4 (Supplementary Table S1), bco2b with primers P5 + P6 and P7 + P8, and bco2 like using primers P9 + P10 and P11 + P12, under the following PCR conditions using liver cDNA as template; 7 min at 94 °C, 94 °C for 30 s, 59 °C for 30 s and 72 °C for 1 min, repeated to a total of 40 cycles, and final extension at 72 °C for 10 min. The salmon bcmo1 gene was amplified with primers P13 + P14 and P15 + P16 using the PCR conditions described above. Amplification of the salmon bcmo1 like sequence was carried out using primers P17 and P18. The resulting fragment of ~1.4 kb was cloned into the pGEM®-T Easy Vector (Promega). Plasmids were sequenced with primers Sp6, M13, P19 and P20. The 3′-end of the salmon bcmo1 like gene was amplified using the FirstChoice® RML-RACE (Ambion) kit, using the gene specific primers P21 and P22. The amplified fragments were sequenced using BigDye® Terminator v3.1 (Applied Biosystems) and 3730 DNA Analyzer (Applied Biosystems). The sequenced bco amplicons were assembled using the Phred-Phrap-Consed package (Ewing and Green, 1998; Ewing et al., 1998; Gordon et al., 1998).

(i.e. column score b 0.93). Each alignment contained gene sequences from ray-finned fish (Actinopterygii) species of the infraclasses Holostei (spotted gar) and Teleostei (zebrafish, Atlantic salmon, stickleback, medaka, Nile tilapia and Tetraodon). In addition, genes from lobe-finned fish (Sarcopterygii) derived species (coelacanth, human, mouse, chicken and Chinese softshell turtle), and a cephalochordate outgroup species (Branchiostoma floridae) that diverged from the vertebrate lineage prior to 2R, was included. Accession numbers for the bco2 and bcmo1 sequences used in the alignments are displayed in supplementary Tables S2 and S3. Phylogenetic trees from amino acid and nucleotide alignments were constructed with maximum likelihood (ML) analyses in MEGA5 (Tamura et al., 2011). The evolutionary model used for ML tree estimation was selected according to the Bayesian Information Criteria calculated with the model test implemented in MEGA5 and 50% consensus trees from 500 bootstrap replications were estimated. 2.6. Gene expression

2.5. Phylogeny Multiple sequence alignments of bcmo1 and bco2 gene families were constructed using the Guidance server with the MAFFT algorithm under the amino acid and codon model (Penn et al., 2010). During the guidance alignment procedure, the alignments were bootstrapped 100 times to identify and remove unreliably aligned amino acids/codons

Relative gene expression was measured in muscle, liver and intestine of 6 Atlantic salmon, using the primers listed in Supplementary Table S1. Two reference genes were evaluated, elongation factor 1alpha (ef1a) and 18S ribosomal RNA. The latter proved to be the most stably expressed, in concurrence with the findings of Jorgensen et al. (2006) and Kortner et al. (2011). Gene expression was determined

A

Fig. 2. Phylogenetic tree of bco2 genes in ray-finned and lobe-finned species, based on nucleotide (A) and amino acid alignment (B). Sequences deviating from known species phylogeny are shown in gray with dashed line. Amphioxus (Branchiostoma floridae) is used as outgroup. (FSGD—fish specific genome duplication).

H. Helgeland et al. / Gene 543 (2014) 268–274

271

B

Fig. 2 (continued).

with qRT-PCR using SYBR®-Green PCR master mix (Applied Biosystems) on Roche Applied Science LightCycler 480, with a 2-step program; initially 95 °C for 10 min, followed by 45 cycles of 95 °C 10 s and 60 °C 1 min. A melt curve analysis was conducted to ensure specificity of the amplicons, and to rule out potential genomic contamination. Two qRT-PCR and two cDNA replicates were used for each sample. Relative expression was calculated using the 2−ΔΔCT method (Livak and Schmittgen, 2001) relative to an average of all measurements within the bco2 and bcmo1 genes, after correcting for PCR efficiency and averaging PCR and RT repeats. One outlier was removed from the bco2 and bcmo1 data. Error bars were calculated from the standard error within the biological replicates of each target gene, and significance testing was performed using one way ANOVA and Tukey HSD pairwise post-hoc test. 3. Results 3.1. Atlantic salmon bco2 and bcmo1 sequences When comparing in-house sequences, ESTs and sequences from the salmon trace archives, we realized that the Atlantic salmon genome contains multiple bco2 and bcmo1 genes. In silico sequence mining and subsequent sequencing revealed 5 beta-carotene oxygenase (bco) genes. The novel bco genes were named according to their phylogenetic relationships, in accordance to the advice from the ZFIN Zebrafish

Nomenclature Committee (personal communication). The three sequenced bco2 cDNAs named bco2a (HF543834), bco2b (HF543835) and bco2 like (HF543836), encoded open reading frames of 1665, 1698 and 1668 bp and corresponding protein products of 574, 585 and 555 amino acids, respectively. Protein alignments show that Bco2a and Bco2b share 72% identical amino acids, while Bco2 like differs more from Bco2a and Bco2b, with an average of 63% protein identity. Sequencing of the two Atlantic salmon bcmo1 amplicons produced open reading frames of 1563 and 1587 bp, encoding corresponding protein products of 520 and 528 aa. The two salmon bcmo1 genes have been named bcmo1 (HF543832) and bcmo1 like (HF543833). Alignment of the predicted salmon Bcmo1 and Bcmo1l proteins shows 54% identical amino acids. All five salmon beta-carotene oxygenases contain the four conserved histidines (His172, His237, His308, His514) and five acidic residues (Asp52, Glu140, Glu314, Glu405, Glu457) described in mouse BCMO1, predicted to be essential for enzyme function (Poliakov et al., 2005). Using the available Atlantic salmon genomic recourses, we recovered partial sequences displaying ~90% identity to the previously described salmon bco genes, proposed to be derived from the salmonid specific genome duplication (4R). However, we were unable to successfully amplify cDNA fragments, suggesting that the proposed 4R derived duplicate genes are not expressed or at low levels. For two of the proposed 4R homeologs (bcmo1II and bcmo1lII), we discovered premature stop codons, caused by indels, within the coding region (data

H. Helgeland et al. / Gene 543 (2014) 268–274

not shown), supporting the hypothesis that they are no longer functional. The partial, suggested 4R sequences were hence not included in the phylogenetic analyses of the proposed functional teleost beta-carotene oxygenases. 3.2. Phylogeny For both bcmo1 and bco2 gene trees the number of teleost paralogs was higher compared to the Sarcopterygii species investigated (i.e. lobed-finned fishes and tetrapods). For bcmo1, amino acid and nucleotide phylogenies produced similar overall topology (Fig. 1 and Supplementary Fig. S1, respectively) with the Sarcopterygii clade nested within two fish-lineages of bcmo1-paralogs. The two different bco2 phylogenies, however, did not concur in overall clade topology. The amino acid based tree shows similar structure as the bcmo1 trees (Fig. 2B), while the nucleotide-based phylogeny places the Sarcopterygii clade as a sister to two teleost-specific bco2 clades (Fig. 2A). Neither of the bco2 trees have high bootstrap support for the conflicting clades. At the individual species level, gene tree topologies mostly concur with known species phylogenies with a few exceptions; placement of spotted gar as a sister to the Sarcopterygii clade in the bco2 amino acid tree (Fig. 2B), placement of zebrafish bco2b outside the teleost bco2b clade (Figs. 2A, B), as well as some inconsistencies in the teleost species relationships within subclades (Figs. 1, S1). The erroneous placement of zebrafish bco2b was resolved when breaking up phylogenetic analyses, using a codon based alignment consisting exclusively of teleost bco2a and bco2b sequences, and a Sarcopterygii outgroup (Supplementary Fig. S2). 3.3. Gene expression of Atlantic salmon bco2 and bcmo1 genes Gene expression of Atlantic salmon bco2a, bco2b and bco2 like was measured in muscle, liver and intestine, using qRT-PCR (Fig. 3). The overall expression of bco2 like is considerably higher than for the other subtypes. Bco2 like show highest expression levels in liver, closely followed by intestine, and the muscle expression is significantly lower than both the intestine (P = 6.3E−05) and liver (P = 7.4E−05). No significant differences between tissues were found in bco2a and bco2b, and the overall expression is low. Gene expression analysis of bcmo1 and bcmo1l revealed distinct tissue specific expression profiles (Fig. 4). In general, bcmo1l exhibits higher expression levels than bcmo1, particularly for liver and intestine. Bcmo1l shows highest expression in intestine, closely followed by liver, and considerably lower in muscle. The muscle expression was significantly lower than both the liver (P = 0.0043) and intestine (P = 0.0039). The expression level of bcmo1 is also highest in intestine, but the intestinal expression is significantly higher than what observed in both the liver (P = 0.0016) and muscle (P = 0.0004). 4. Discussion

we did not find evidence for more than two bcmo1 paralogs in the investigated teleost species. Other possible explanations for the clustering of piscine Bcmo1 with the Sarcopterygii orthologs could be convergent evolution or rapid functional diversification of bcmo1l genes. However, the evolutionary relationships observed in our study; bcmo1 paralogs arising prior to the divergence of ray-finned and lobe-finned fish, are consistent with the findings of Albalat et al. (2011) and the Ensemble Gene Tree ENSGT00500000044783, showing the phylogeny of the extended RPE65 gene family. Our phylogenetic analyses of the bco2 family did not produce consistent topologies to support a single model of Bco2-paralog evolution. The results from the nucleotide-based phylogeny support two teleost specific duplications (Fig. 2A), while the amino acid based tree suggests that teleosts have retained an ancient additional Bco2 copy that has been lost from the Sarcopterygii lineage (Fig. 2B). However, although the two bco2 phylogenies deviate, several lines of evidence can be helpful to assess the probability of these two different evolutionary scenarios. Only a single spotted gar bco2 sequence is found, which suggests that the bco2 duplications either has happened subsequent to the garteleost clade divergence, or alternatively (if the amino acid tree is correct) that both gar and the Sarcopterygii lineage have independently lost a bco2 copy subsequent to their divergence from the teleosts. Hence, the origin of bco2l due to a gene or segment duplication that occurred after the teleost divergence, represents a more parsimonious model of evolution, which is supported by Albalat et al. (2011) and the Ensemble Gene Tree ENSGT00500000044783. Assuming this latter model, we would expect an additional bco2l paralog derived from the FSGD. However, we have not found evidence for a fourth bco2 paralog in the investigated teleost species suggesting that this hypothesized teleost bco2l paralog has lost its function or been lost from the genome. Phylogenetic analyses of the bco2 gene family indicate that bco2a and bco2b arose from a duplication in the teleost lineage, possibly the fish specific genome duplication, FSGD or 3R. Chromosomal locations of bco2a and bco2b genes are known in medaka, Tetraodon and zebrafish

4

3

RelExp

272

tissue muscle liver intestine

2

4.1. The evolutionary history of the piscine beta-carotene oxygenases The mammalian family of carotenoid oxygenases contains two members, BCMO1 and BCO2. The teleost enzyme family, however, includes 5 members: two bcmo1 genes and three bco2 genes. Bcmo1 and bcmo1 like, appear to be the result of an ancient duplication event that took place prior to the divergence of ray-finned and lobe-finned fish (Fig. 1 and Supplementary Fig. S2). This is supported by the presence of two spotted gar bcmo1 paralogs, an Actinopterygii species that diverged from the teleosts before the FSGD (Amores et al., 2011). In the coelacanth, mammalian, avian and reptile species compared, we found no evidence of a second BCMO1 paralog, indicating that the hypothesized Sarcopterygii paralog has been subjected to non-functionalization and lost, a fate that is shared with most duplicated genes (Lynch and Conery, 2000). In addition, given that bcmo1 and bcmo1l arose prior to the divergence of teleost fish, the expected paralogs from the FSGD appear also to have been lost, as

1

0 bco2a

bco2b

bco2

genes Fig. 3. Relative gene expression of bco2a, bco2b and bco2 like in the muscle, liver and intestine of Atlantic salmon. Values are expressed relative to the mean of all measurements in the three paralogs. There are 6 biological replicates within the tissue groups (except liver, n = 5). Error bars are calculated from the standard error within the tissue groups.

H. Helgeland et al. / Gene 543 (2014) 268–274

2.5

RelExp

2.0

tissue muscle liver intestine

1.5

1.0

0.5

0.0 bcmo1

genes

bcmo1 like

Fig. 4. Relative gene expression of bcmo1 and bcmo1 like in the muscle, liver and intestine of Atlantic salmon. Values are expressed relative to the mean of all measurements in both paralogs. There are 6 biological replicates within the tissue groups (except intestine, n = 5). Error bars are calculated from the standard error within the tissue groups.

(Supplementary Table S2), and this could shed light on their origin. The medaka bco2a ortholog is positioned on chromosome 13 and the bco2b ortholog on chromosome 14. The medaka chromosomes 13 and 14, consist in large parts of ancestral chromosome segments from the duplicated reconstructed ancestral chromosome ‘h’, proposed to stem from the FSGD (Kasahara et al., 2007). In Tetraodon, the ancestral chromosome segments from ‘h’ are mainly found on Tetraodonchromosomes 7 and 16 and indeed these chromosomes contain the Tetraodon orthologs bco2a (Tni16) and bco2b (Tni7). The zebrafish genome has undergone extensive rearrangements, and ancestral chromosome segments from ‘h’ are found on Dre10, 21, 5 and 15, as are the zebrafish bco2a and bco2b orthologs, located on Dre15 and 5. Hsa11q22–23, containing the human BCO2 gene is syntenic with chromosomal regions of both zebrafish chromosomes 15 and 5 (Auer et al., 2007), and the human chromosome 11 contains ancestral chromosome segments from the reconstructed chromosome ‘h’ (Kasahara et al., 2007). We therefore suggest that the youngest node in Figs. 2A and B separating bco2a and bco2b represents the FSGD. 4.2. Functional divergence of the beta-carotene oxygenases While the majority of duplicated genes are lost within a few million years, subsequent to a gene or genome duplication event (Lynch and Conery, 2000), teleost fish appear to have retained an extended family of beta-carotene oxygenases. Although subfunctionalization can occur under near-neutral processes (Force et al., 2005), once established there will be a selective pressure to retain both duplicates (Force et al., 1999). Gene expression levels for the 5 Atlantic salmon beta-carotene oxygenases were measured in the intestine, liver and muscle, tissues exerting key functions in the uptake, metabolism and storage of carotenoids in salmonids (Page and Davies, 2006; Storebakken and No, 1992). The differences in overall and tissue specific gene expression, and the structural divergence of the bco2 and bcmo1 paralogs, suggests

273

that the bco genes have been subjected to functional divergence. Bcmo1 and bcmo1 like are both highly expressed in Atlantic salmon intestine and liver, tissues showing high beta-carotene oxygenase activity in mammals (Parker, 1996; Yeum and Russell, 2002). The beta-carotene cleaving property of the zebrafish Bcmo1 protein (cDNA AJ290390) was confirmed by the presence of retinal after co-expressing the enzyme in a beta-carotene synthesizing E. coli strain (Lampert et al., 2003). Interestingly, the knockdown morpholino model studied by Lampert and colleagues revealed that Bcmo1 plays a vital role during zebrafish development. The Bcmo1-deficient embryos and larvae displayed indicators of impaired retinoic acid-dependent developmental processes, indicating that the bcmo1l product could not compensate for the RA deficiency. The salmon bcmo1l gene shows considerably higher expression in both the intestine and liver compared to bcmo1, and the tissue specific profile is particularly divergent in the liver (Fig. 4). Based on the high expression levels we hypothesize that the bcmo1 like gene has a functional gene product. Although it appears to lack pro-vitamin A cleaving properties, it still may play an important role in carotenoid metabolism. Comparing the salmon bco2 genes, bco2l shows vastly higher expression than to the two paralogs (Fig. 3). Bco2a and bco2b display low overall expression in salmon, and no significant differences between tissues. Asymmetric divergence, where one duplicate has lost more functions than the other(s), is not uncommon and is proposed to increase robustness to deleterious mutations (Wagner, 2002). The high expression of Bco2 like, particularly in the intestine and liver, indicates that salmon Bco2 like is functional and may be the quantitatively most important beta-carotene oxygenase 2 in salmon. This is in accordance with the findings of Lobo et al. (2012), where the zebrafish Bco2l ortholog is presented as the sole functional beta-carotene oxygenase 2. There are over 600 naturally occurring carotenoids, and great structural diversity is displayed in marine carotenoids (Liaaenjensen, 1991; Matsuno, 2001). The most structurally diverse carotenoids are found in algae and bacteria, fungi are less diverse, and terrestrial plants have the least variable carotenoids (Johnson and Schroeder, 1996). This appears true for carotenoids in a lacustrine environment as well. Comparing fossil pigments in woodland, swamps and lakes, Jon and Gorham (1973) found that carotenoid abundance and diversity increased with the degree of soil waterlogging, with lakes showing the highest richness and variation. It is thus possible that the abundance and structural diversity of aquatic carotenoids may have favored specialized carotenoid cleavage enzymes in fish. In conclusion, the differences in overall and tissue specific gene expression, and the divergence of the bco2 and bcmo1 paralogs, as revealed by sequence alignments and phylogenetic analyses, suggest that the beta-carotene oxygenase subtypes may have evolved functional divergences beneficial to the teleost lineage. We hypothesize that the retention of an extended family of beta-carotene oxygenases in teleost fish is due to better utilization of the diverse carotenoids in an aquatic environment. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.gene.2014.02.042. Author contributions Hanna Helgeland—concept, design, material, data collection, analysis and interpretation, literature search, writing and critical review. Simen Rød Sandve—data analysis, interpretation and writing of the phylogeny section, and critical review. Jacob Seilø Torgersen—data collection, analysis, interpretation and writing related to the bcmo1 like sequence. Mari Kyllesø Halle—data collection and analysis of the bcmo1 like sequence. Hilde Sundvold—design, supervision and interpretation. Stig Omholt—concept, resources and critical review. Dag Inge Våge— concept, design, supervision, resources, material, interpretation and critical review.

274

H. Helgeland et al. / Gene 543 (2014) 268–274

All authors have read and approved the final manuscript. The author contributions follow guidelines from the Council of Science Editors (CSE) and the Vancouver Convention. http://openwetware.org/wiki/ Authorship. Competing interests statement The authors declare no competing interests. Acknowledgments Thanks to Simon Taylor for assembling the de novo draft salmon genome. We also thank Amy Singer and John Postlethwait at the Zebrafish Nomenclature Committee (ZNC) for valuable advice and discussion concerning the naming of the salmon bco genes. Funding for the Atlantic salmon sequencing project was made by The International Cooperation Project to Sequence the Atlantic Salmon Genome (ICSASG: http:// www.genomebc.ca/partners/international-collaborators/internationalcooperation-to-sequence-the-atlantic-salmon-geno/). References Albalat, R., Brunet, F., Laudet, V., Schubert, M., 2011. Evolution of retinoid and steroid signaling: vertebrate diversification from an amphioxus perspective. Genome Biology and Evolution 3, 985–1005. Allendorf, F., Thorgaard, G., 1984. Tetraploidy and the evolution of salmonid fishes. In Evolutionary Genetics of Fishes. N. Edited by Turner BJ. Plenum Press, pp. 1–46. Amengual, J., Lobo, G.P., Golczak, M., Li, H.N.M., Klimova, T., Hoppel, C.L., Wyss, A., Palczewski, K., von Lintig, J., 2011. A mitochondrial enzyme degrades carotenoids and protects against oxidative stress. The FASEB Journal: Official Publication of the Federation of American Societies for Experimental Biology 25 (3), 948–959. Amores, A., Catchen, J., Ferrara, A., Fontenot, Q., Postlethwait, J.H., 2011. Genome evolution and meiotic maps by massively parallel DNA sequencing: spotted gar, an outgroup for the teleost genome duplication. Genetics 188 (4) (799-U79). Auer, R.L., Riaz, S., Cotter, F.E., 2007. The 13q and 11q B-cell chronic lymphocytic leukaemia-associated regions derive from a common ancestral region in the zebrafish. British Journal of Haematology 137 (5), 443–453. Christoffels, A., Koh, E.G.L., Chia, J.M., Brenner, S., Aparicio, S., Venkatesh, B., 2004. Fugu genome analysis provides evidence for a whole-genome duplication early during the evolution of ray-finned fishes. Molecular Biology and Evolution 21 (6), 1146–1151. Davidson, W., Koop, B., Jones, S., Iturra, P., Vidal, R., Maass, A., Jonassen, I., Lien, S., Omholt, S., 2010. Sequencing the genome of the Atlantic salmon (Salmo salar). Genome Biology 11 (9), 403. Ewing, B., Green, P., 1998. Base-calling of automated sequencer traces using phred. II. Error probabilities. Genome Research 8 (3), 186–194. Ewing, B., Hillier, L., Wendl, M.C., Green, P., 1998. Base-calling of automated sequencer traces UsingPhred.I. Accuracy assessment. Genome Research 8 (3), 175–185. Force, A., Lynch, M., Pickett, F.B., Amores, A., Yan, Y.-l, Postlethwait, J., 1999. Preservation of duplicate genes by complementary, degenerative mutations. Genetics 151 (4), 1531–1545. Force, A., Cresko, W.A., Pickett, F.B., Proulx, S.R., Amemiya, C., Lynch, M., 2005. The origin of subfunctions and modular gene regulation. Genetics 170 (1), 433–446. Gordon, D., Abajian, C., Green, P., 1998. Consed: a graphical tool for sequence finishing. Genome Research 8 (3), 195–202. Jaillon, O., Aury, J.M., Brunet, F., Petit, J.L., Stange-Thomann, N., Mauceli, E., Bouneau, L., Fischer, C., Ozouf-Costaz, C., Bernot, A., et al., 2004. Genome duplication in the teleost fish Tetraodon nigroviridis reveals the early vertebrate proto-karyotype. Nature 431 (7011), 946–957. Johnson, E., Schroeder, W., 1996. Microbial carotenoids downstream processing biosurfactants carotenoids. Advances in Biochemical Engineering/Biotechnology, vol. 53. Springer, Berlin/Heidelberg, pp. 119–178. Jon, E.S., Gorham, E., 1973. A comparison of the abundance and diversity of fossil pigments in wetland peats and woodland humus layers. Ecology 54 (3), 605–611. Jorgensen, S.M., Kleveland, E.J., Grimholt, U., Gjoen, T., 2006. Validation of reference genes for real-time polymerase chain reaction studies in Atlantic salmon. Marine Biotechnology 8 (4), 398–408. Kasahara, M., Naruse, K., Sasaki, S., Nakatani, Y., Qu, W., Ahsan, B., Yamada, T., Nagayasu, Y., Doi, K., Kasai, Y., et al., 2007. The medaka draft genome and insights into vertebrate genome evolution. Nature 447 (7145), 714–719. Kiefer, C., Hessel, S., Lampert, J.M., Vogt, K., Lederer, M.O., Breithaupt, D.E., von Lintig, J., 2001. Identification and characterization of a mammalian enzyme catalyzing the

asymmetric oxidative cleavage of provitamin A. Journal of Biological Chemistry 276 (17), 14110–14116. Kim, Y.-S., Yeom, S.-J., Oh, D.-K., 2011. Production of β-apo-10′-carotenal from β-carotene by human β-carotene-9′,10′-oxygenase expressed in E. coli. Biotechnology Letters 33 (6), 1195–1200. Kortner, T.M., Valen, E.C., Kortner, H., Marjara, I.S., Krogdahl, A., Bakke, A.M., 2011. Candidate reference genes for quantitative real-time PCR (qPCR) assays during development of a diet-related enteropathy in Atlantic salmon (Salmo salar L.) and the potential pitfalls of uncritical use of normalization software tools. Aquaculture 318 (3–4), 355–363. Lampert, J.M., Holzschuh, J., Hessel, S., Driever, W., Vogt, K., von Lintig, J., 2003. Provitamin A conversion to retinal via the beta, beta-carotene-15,15′-oxygenase (bcox) is essential for pattern formation and differentiation during zebrafish embryogenesis. Development 130 (10), 2173–2186. Liaaenjensen, S., 1991. Marine carotenoids—recent progress. Pure and Applied Chemistry 63 (1), 1–12. Lien, S., et al., 2001. “A dense SNP-based linkage map for Atlantic salmon (Salmo salar) reveals extended chromosome homeologies and striking differences in sex-specific recombination patterns.”. Bmc Genomics 12 (1), 615. Lindqvist, A., Andersson, S., 2002. Biochemical properties of purified recombinant human beta-carotene 15,15′-monooxygenase. Journal of Biological Chemistry 277 (26), 23942–23948. Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data using realtime quantitative PCR and the 2−ΔΔCT method. Methods 25 (4), 402–408. Lobo, G.P., Amengual, J., Palczewski, G., Babino, D., von Lintig, J., 2011. Carotenoidoxygenases: key players for carotenoid function and homeostasis in mammalian biology. Biochimica et Biophysica Acta 1821 (1), 78–87. Lobo, G.P., Isken, A., Hoff, S., Babino, D., von Lintig, J., 2012. BCDO2 acts as a carotenoid scavenger and gatekeeper for the mitochondrial apoptotic pathway. Development 139 (16), 2966–2977. Lynch, M., Conery, J.S., 2000. The evolutionary fate and consequences of duplicate genes. Science 290 (5494), 1151–1155. Matsuno, T., 2001. Aquatic animal carotenoids. Fisheries Science 67 (5), 771–783. Mein, J.R., Dolnikowski, G.G., Ernst, H., Russell, R.M., Wang, X.-D., 2011. Enzymatic formation of apo-carotenoids from the xanthophyll carotenoids lutein, zeaxanthin and [beta]-cryptoxanthin by ferret carotene-9′,10′-monooxygenase. Archives of Biochemistry and Biophysics 506 (1), 109–121. Page, G.I., Davies, S.J., 2006. Tissue astaxanthin and canthaxanthin distribution in rainbow trout (Oncorhynchus mykiss) and Atlantic salmon (Salmo salar). Comparative Biochemistry and Physiology. Part A, Molecular & Integrative Physiology 143 (1), 125–132. Parker, R.S., 1996. Carotenoids.4. Absorption, metabolism, and transport of carotenoids. The FASEB Journal: Official Publication of the Federation of American Societies for Experimental Biology 10 (5), 542–551. Penn, O., Privman, E., Ashkenazy, H., Landan, G., Graur, D., Pupko, T., 2010. GUIDANCE: a web server for assessing alignment confidence scores. Nucleic Acids Research 38 (Suppl. 2), W23–W28. Poliakov, E., Gentleman, S., Cunningham, F.X., Miller-Ihli, N.J., Redmond, T.M., 2005. Key role of conserved histidines in recombinant mouse beta-carotene 15,15′monooxygenase-1 activity. Journal of Biological Chemistry 280 (32), 29217–29223. Simões-costa, M.S., Azambuja, A.P., Xavier-Neto, J., 2008. The search for non-chordate retinoic acid signaling: lessons from chordates. Journal of Experimental Zoology. Part B, Molecular and Developmental Evolution 310B (1), 54–72. Smith, J.J., Kuraku, S., Holt, C., Sauka-Spengler, T., Jiang, N., Campbell, M.S., Yandell, M.D., Manousaki, T., Meyer, A., Bloom, O.E., et al., 2013. Sequencing of the sea lamprey (Petromyzon marinus) genome provides insights into vertebrate evolution. Nature Genetics 45 (4), 415–421. Storebakken, T., No, H.K., 1992. Pigmentation of rainbow trout. Aquaculture 100 (1–3), 209–229. Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., Kumar, S., 2011. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Molecular Biology and Evolution 28 (10), 2731–2739. Untergasser, A., Nijveen, H., Rao, X., Bisseling, T., Geurts, R., Leunissen, J.A.M., 2007. Primer3Plus, an enhanced web interface to Primer3. Nucleic Acids Research 35, W71–W74. von Lintig, J., Vogt, K., 2000. Filling the gap in vitamin A research—molecular identification of an enzyme cleaving beta-carotene to retinal. Journal of Biological Chemistry 275 (16), 11915–11920. Wagner, A., 2002. Asymmetric functional divergence of duplicate genes in yeast. Molecular Biology and Evolution 19 (10), 1760–1768. Wyss, A., Wirtz, G., Woggon, W.D., Brugger, R., Wyss, M., Friedlein, A., Bachmann, H., Hunziker, W., 2000. Cloning and expression of beta, beta-carotene 15,15′dioxygenase. Biochemical and Biophysical Research Communications 271 (2), 334–336. Yeum, K.J., Russell, R.M., 2002. Carotenoid bioavailability and bioconversion. Annual Review of Nutrition 22, 483–504.