Comparative analysis of zebrafish nos2a and nos2b genes

Gene 445 (2009) 58–65

Contents lists available at ScienceDirect

Gene j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / g e n e

Comparative analysis of zebrafish nos2a and nos2b genes Sandrine Lepiller, Nathalie Franche, Eric Solary, Johanna Chluba ⁎, Véronique Laurens Inserm UMR 866, University of Burgundy, Institut Fédératif de Recherche Santé STIC, 21000 Dijon, France

a r t i c l e

i n f o

Article history: Received 26 September 2008 Received in revised form 22 May 2009 Accepted 26 May 2009 Available online 6 June 2009 Received by M. Schartl Keywords: Inducible nitric oxide synthase Gene duplication Synteny Evolution Teleostei

a b s t r a c t Nitric oxide synthase (NOS) produces nitric oxide (NO) from arginine. Three NOS isoforms have been identified in mammals, namely a neuronal (NOS1), an inducible (NOS2) and an endothelial (NOS3) enzyme. In zebrafish genome, one nos1 gene and two nos2 genes (nos2a and nos2b) were observed. We cloned zebrafish nos2a cDNA and compared nos2a and nos2b sequences, expression and inducibility. When analyzed by reverse transcription-PCR, the expression of nos2a remained very low during initial development, then increased at 96 hpf, while nos2b was expressed from 6 hpf and subsequently remained stable. Expression of nos2a is detected in the head, eye and gut regions by WISH experiments performed at 48, 72 and 96 hpf larvae. In adults, nos2a expression varies from one tissue to another whereas nos2b is expressed in all studied tissues. Both nos2 isoforms can be induced by pro-inflammatory or mechanical stresses (tissue injury). In vitro as in vivo stimulations with Poly I:C and lipopolysaccharides (LPS) enhanced more dramatically nos2a than nos2b expression. After tail transection in 4 dpf larvae a strong increase of nos2a and nos2b expression was evidenced in the regeneration site, skin cells and for the nos2b gene in neuromasts. Phylogenetic and syntenic analyses show that nos2b gene was associated with syntenic genes identified for nos2 genes in vertebrate. This is not the case for the nos2a gene, despite zebrafish nos2a presenting the inducible property of a classical vertebrate nos2 isoform. A myristoylation consensus site was detected at the N-terminal extremity of zebrafish Nos2b, a property shared with mammal NOS3 isoforms. Thus, the evolution of nos2 genes in zebrafish provides a typical example of gene divergence after duplication. © 2009 Elsevier B.V. All rights reserved.

1. Introduction Nitric oxide synthases (NOS) catalyze the formation of nitric oxide (NO) by converting L-arginine into citrulline. In mammals, three NOS have been identified: two of these enzymes are constitutively expressed while the expression of the third one is inducible. The two constitutive NOS are neuronal (nNOS or NOS1) and endothelial (eNOS or NOS3) enzymes. The inducible NOS (iNOS or NOS2) is expressed in response to pro-inflammatory cytokines and/or bacterial products such as lipopolysaccharides (LPS). The transcriptional induction of nos2 gene is responsible for a sustained production of high NO levels (Alderton et al., 2001). NO produced by NOS2 contributes to an appropriate immune defense against pathogens but is also involved in the pathophysiology of malignant, autoimmune, inflammatory and neurodegenerative diseases (Kroncke et al., 1997, 2000; Jeannin et al., 2008). NOS2 has been described in mammals and birds (Lin et al, 1996). In bony fishes, NOS2 protein has been detected (Shin et al., 2000) and Abbreviations: NO, nitric oxide; NOS, nitric oxide synthase; WISH, whole mount in situ hybridization; Poly I:C, polyinosine-polycytidylic acid; LPS, lipopolysaccharide; DAF-FM DA, 4-amino-5-mehtylamino-2’-7’-difluorofluorescein diacetate; NAPDH, reduced form of nicotinamide adenine dinucleotide phosphate; FMN, flavin mononucleotide; FAD, flavin adenine dinucleotide. ⁎ Corresponding author. Inserm UMR 866, 6 Boulevard Gabriel, Dijon F-21000, France. Tel./fax: +33 3 80 39 62 23. E-mail address: [email protected] (J. Chluba). 0378-1119/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.gene.2009.05.016

genes encoding nos2 homologs have been cloned in goldfish, Carassius auratus (Laing et al., 1996), rainbow trout, Oncorhynchus mykiss (Wang et al., 2001) and carp, Cyprinus carpio (Saeij et al., 2000). NOS2 has also been identified in a cartilaginous fish, the small spotted catshark, Scyliorhinus canicula (Reddick et al., 2006). Upon bacterial, viral or parasitic infection, LPS or Poly I:C stimulation, fish nos2 expression increases, which may lead to an increase of NO production, as demonstrated in carp phagocytes (Campos-Perez et al., 2000; Bridle et al., 2006; Gonzalez et al., 2007). Using a fluorescent NO specific probe, the diaminofluorescein DAF-FM-DA, we previously reported different stressful conditions, where an enhancement of NO production can be observed in zebrafish larvae, as in the regeneration site after tail transection (Lepiller et al., 2007). The characterization of nos isoforms induced in such condition remained to be elucidated and we focus our work on the identification of inducible zebrafish Nos isoforms. In zebrafish, Danio rerio, the first nos gene described, located on chromosome 5, was the ortholog of mammalian nos1 gene (Holmqvist et al., 2000; Poon et al., 2003). In the Ensembl database (www. ensembl.org/), the zebrafish genome includes two nos2 genes, nos2a and nos2b. The latter gene has been recently cloned and its expression pattern was described (Poon et al., 2008). Located on chromosome 15, this nos2b gene is constitutively expressed in embryonic zebrafish, especially in tissues surrounding the oral cavity. We here report the cloning of zebrafish nos2a cDNA whose sequence, genomic

S. Lepiller et al. / Gene 445 (2009) 58–65

59

Zebrafish embryos and larvae were grown in E3 medium (5 mmol/L NaCl, 0.17 mmol/L KCl, 0.4 mmol/L CaCl2, and 0.16 mmol/L MgSO4), containing 0.003% 1-phenyl-2-thiourea (PTU; Sigma, St. Louis, MO, USA) to block pigment development for WISH experiments, and staged according to Kimmel et al. (1995). Complete transection of the tail was performed with a sterile scalpel, midway between the anal fin and the base of the caudal fin. Adult zebrafish (D. rerio) (ZIRC, Oregon, USA) were maintained as described (Westerfield, 2000). Animal care and experimentation were done in compliance with French and European laws (Authorization no. 21-CC-EL-011 from 09/02/2005).

(NJplot software) was used to construct a phylogenic tree (Saitou and Nei, 1987). Bootstrap analysis was made on 1000 sampling steps. The accession numbers of the sequences used are as follows: human NOS1 (D16408), mouse NOS1 (D14552), rat NOS1 (U67309), Xenopus NOS1 (NM001085686), fugu NOS1 (AF380137), medaka NOS1 (AB163430), zebrafish Nos1 (AY211528), human NOS2 (L09210), mouse NOS2 (M87039), rat NOS2 (D44591), chicken NOS2 (U46504), carp NOS2 (AJ242906), trout NOS2 (AJ300555), goldfish NOS2 form a (AY904362) and NOS2form b (AY904363), catshark NOS2 (AY904361), human NOS3 (M93718), mouse NOS3 (U53142), rat NOS3 (NM_021838), pig NOS3 (U59924), and Drosophila NOS (U25117). Other predicted sequences via Ensembl peptide ID Ciona NOS (ENSCINP00000005536), Xenopus NOS2 (ENSXETT00000027328), and Xenopus NOS3 (ENSXETP00000025059) were used. Genomic localization of nos genes and identification of syntenic genes were performed by BLAST analyses and genome comparison on Ensembl release 47 and NCBI assembly genome data. The versions used include the human NCBI 36, the mouse NCBI m37, the chicken 2.1 release, the Xenopus tropicalis 4.1 version, the version 7 zebrafish (Zv7) and the version 2.0 Ciona intestinalis genome assemblies.

2.2. Cloning of the nos2 cDNA

2.4. RT-PCR analysis of nos2a and nos2b mRNA

The search for sequences with similarities to the human and mouse nos2 genes was performed by using the Ensembl Danio rerio database (http://www.ensembl.org). RT-PCR primers used to amplify the complete coding region of the genes are summarized in Table 1. The PCR program was as follows: initial denaturation at 94 °C for 4 min, 35 cycles of amplification [94 °C for 1 min, 59 °C or 69 °C for 1 min, depending of the studied nos2 gene, and 72 °C for 3 min] and a final extension step at 72 °C for 10 min. Amplified PCR products were gel purified, cloned into pPCR®-XL-TOPO (Invitrogen, Cergy Pontoise, France) and sequenced.

The expression of the two zebrafish nos2 genes was studied by RT-PCR on total RNA obtained from embryos, larvae and adult tissues. Total RNA was isolated using the Trizol Reagent (Invitrogen) and reverse transcribed using the SuperScript™ First-Strand Synthesis for RT-PCR kit (Invitrogen). The primers used in these experiments are summarized in Table 1. The PCR program was as follows: initial denaturation for 4 min at 94 °C, followed by 35 cycles [94 °C for 1 min, 60 °C for 1 min and 72 °C for 1 min] and a final extension at 72 °C for 10 min. The PCR products were detected by electrophoresis in 1.5% agarose gel.

2.3. Sequences, phylogenetic and syntenic analyses

2.5. Whole-mount in situ hybridization

Molecular mass of deduced amino acids sequences was predicted with the Compute pI/Mw tool (http://www.expasy.ch). The prediction of N-myristoylation sequence was obtained using NMT— the MYR Predictor web server (http://mendel.imp.ac.at/myristate/ SUPLpredictor.htm). The CLUSTALW multiple sequence alignment program was used to determine homologies between amino sequences of zebrafish and other animal NOS and to perform sequence alignments (Chenna et al., 2003). The neighbor-joining method

The 918 bp and 949 bp PCR-fragments of zebrafish nos2a and nos2b cDNA, respectively, were subcloned into pCR4-TOPO vector (Invitrogen). PCR with M13 reverse and M13 forward primers were performed on recombinant plasmids. These PCR products were used as template to synthesize digoxigenin (DIG)-labeled riboprobes using T3 and T7 RNA polymerases (DIG-labeling kit; Roche Lab., Indianapolis, IN). The specificity of antisense probes was verified by running control experiments with sense probes. Whole-mount in situ hybridization (WISH) experiments were performed as described (see: http://zfin. org/zf_info/zfbook/chapt9/9.8.html) on embryos fixed at 48, 72 and 98 h post-fertilization (hpf). Hybridization was performed in 50% formamide and DIG-labeled probes were detected with anti-fluorescein alkaline phosphatase-coupled Fab fragments (1:5000) and NBT/BCIP (Roche). Embryos were cleared in graduate concentration of glycerol up to 80% before observations. At least 15 embryos were examined and demonstrated similar expression patterns.

localization, expression along development and in adult tissues was compared to that of nos2b gene. We evaluated in vitro and in vivo induction of both genes in cells exposed to LPS and poly I:C and performed phylogeny and synteny analyses to discuss nos gene evolution in vertebrates. 2. Materials and methods 2.1. Maintenance of fishes

Table 1 Oligonucleotide primers used for cloning zebrafish cDNAs and analyzing their expression by RT-PCR and WISH. Name

Sequence

For cDNA cloning nos2a 5′-GACCAGATAACCACTGCTCTG 5′-GATGGGAAAAAACCATCTGAATC nos2b 5′-ATGTTCCTGTGGTCAGACAGT 5′-TGAGCACTAGTATCCGAAGATG For RT-PCR analysis nos2a 5′-GTGTTCCCTCAGAGAACAGAT 5′-GATCAGTCCTTTGAAGCTGAC nos2b 5′-GATAGACGGCACATTATTAGGA 5′-GAAGCAACAGTTCATGATGCC ifn1 5′-TCTGCGTCTACTTGCGAATG 5′-CGTTTCGTTGATGATTGCTG β-actin 5′-CAGCCATGGATGAGGAAATC 5′-TCACACCATCACCAGAGTCC For WISH analysis nos2a 5′-CAGGACAGTTGCACTCTGGA 5′-CAGTAGGGTCATGGGGCTAA nos2b 5′-GGTTGTTTGCATGGAGGACT 5′-CTCCAGCACCTCAAGGAAAG

Type

Length of amplicons

Forward Reverse Forward Reverse

3382 bp

Forward Reverse Forward Reverse Forward Reverse Forward Reverse

800 bp

Forward Reverse Forward Reverse

3429 bp

2.6. Stimulation of ZFL and ZF4 cell lines

880 bp 560 bp 483 bp

918 bp 949 bp

Zebrafish liver cells ZFL (CRL-2643, ATCC, Rockville, MD, USA) were grown at 28 °C in a medium containing a combination of 50% L-15, 35% Dulbecco's minimum essential medium (DMEM) and 15% F-12 with sodium selenite supplemented with 10% fetal bovine serum (FBS), 15 mM HEPES, 0.01 mg/mL insulin, 100 U/mL penicillin, 100 μg/mL streptomycin and 250 ng/mL amphotericin B. The zebrafish embryo fibroblast cell line ZF4 (CRL-2050, ATCC) was grown in a mixture of F-12 and DMEM (1:1) supplemented with 10% FBS, 10 mM HEPES, 100 U/mL penicillin, 100 μg/mL streptomycin, and 250 ng/mL amphotericin B. ZFL and ZF4 cells were stimulated with 50 μg/mL Poly I:C (Sigma, St. Louis, MO, USA) or 50 μg/mL Salmonella typhimurium

60

S. Lepiller et al. / Gene 445 (2009) 58–65

lipopolysaccharides (LPS) (Sigma) and harvested at 8 and 20 h post-treatment. 2.7. Challenge of adult zebrafish with LPS Adult zebrafish were injected with 50 μL of S. typhimurium LPS (5 μg in PBS) or 50 μL PBS (control) in ventral muscle and euthanized 6 h later to collect the tissue at the injection site for RT-PCR analysis. 3. Results 3.1. Identification and phylogenetic analysis of nos2 genes in zebrafish Bioinformatic analysis of zebrafish genome indicated two potential nos2 genes whose complete coding region was cloned and sequenced: the 3382 bp nos2a cDNA (accession number AM749801) encodes a 1079 amino acid protein with a 121.8 kDa predicted molecular weight and the 3435 bp nos2b cDNA (accession number AM749802) encodes a 1077 amino acid protein with a 122 kDa predicted molecular weight.

The latest sequence shares 98% amino acid identity with the deduced amino acid sequence of recently described Nos2b sequence (GenBank accession no. EU332350) (Poon et al., 2008). The percentage of amino acid identity between the Nos2a and Nos2b proteins is 68% (Fig. 1). When compared to other NOS sequences, the predicted amino acid sequences of the two zebrafish Nos2 share the highest homologies to bony fish NOS2 (67–84% amino acid identity to goldfish, carp and trout NOS2) and to other vertebrate NOS2 (55–60% amino acid identities) in comparison to constitutive vertebrate NOS3 and NOS1 (48–51% amino acid identities) (Table 2). All important functional domains identified in vertebrate NOS2 were observed in the two zebrafish Nos2 (Fig. 1). The two enzymes contain the cofactor-binding sites for heme, tetrahydrobiopterin, calmodulin, flavin mononucleotide (FMN), flavin adenine dinucleotide (FAD) pyrophosphate, FAD isoalloxazine, NADPH ribose, NADPH adenine, and a conserved Cterminal sequence for NADPH binding (Fig. 1). Surprisingly, Nos2b amino acid sequence analysis revealed an N-terminal myristoylation consensus sequence that is characteristic of NOS3 in mammals. A nos3 gene has never been identified in fishes and we could not find

Fig. 1. Comparison of the deduced amino acid sequences of zebrafish Nos2a and Nos2b. The sequences were aligned using ClustalW. Identical amino acids are indicated by asterisks, conservative substitutions are shown by colon, and semi-conservative substitutions by dots. Gaps in the sequence are represented by dashes. Predicted N-myristoylation site, the conserved cofactor-binding sites for heme, tetrahydrobiopterin, calmodulin, flavine mononucleotide (FMN), flavine adenine dinucleotide (FAD) pyrophosphate, FAD isoalloxazine, NADPH ribose, NADPH adenine, and the C-terminal conserved sequence for NADPH binding are boxed in grey.

S. Lepiller et al. / Gene 445 (2009) 58–65 Table 2 Homologies of zebrafish Nos2a and Nos2b to other known NOS sequences. Isoform

Amino acid identity to zebrafish Nos2a (%)

Amino acid identity to zebrafish Nos2b(%)

Human NOS1 Mouse NOS1 Rat NOS1 Chicken NOS1 Xenopus NOS1 Zebrafish NOS1 Human NOS3 Mouse NOS3 Rat NOS3 Pig NOS3 Xenopus NOS3 Human NOS2 Mouse NOS2 Rat NOS2 Chicken NOS2 Xenopus NOS2 Carp NOS2 Trout NOS2 Goldfish NOS2a Goldfish NOS2b Catshark NOS2 Ciona NOS Drosophila NOS

50.6 50.8 49.3 51.1 50.9 51.7 49.1 50.3 50.0 49.0 50.2 58.7 58.3 58.0 58.2 55.3 69.4 66.8 71.2 71.5 56.8 45.5 43.4

51.1 50.9 48.6 51.2 49.8 51.6 50.1 50.5 50.3 49.1 51.3 60.0 57.9 58.3 58.3 55.3 83.9 67.7 83.6 85.4 56.4 44.9 43.1

any nos3 sequence in fish genome databases. The earliest nos3 homolog sequence was detected in the Xenopus genome but incomplete nucleotide sequence precluded identification of an N-terminal myristoylation consensus sequence. The two nos2 forms found in goldfish share 86.9% nucleotide identities and 90.3% amino acid identities. Interestingly, these goldfish nos2a and nos2b sequences encode proteins with an N-terminal myristoylation consensus sequence. Conversely, using Ensembl database, we failed to identify nos2 genes in fugu, medaka, stickleback and tetraodon genomes. Phylogenetic analysis of the predicted NOS2 amino acid sequences indicates that zebrafish Nos2a and Nos2b formed a cluster with other vertebrate NOS2. Bony fish NOS2 forms a branch that is distinct from other fish NOS2 sequences (Fig. 2). A NOS sequence with a putative N-myristoylation site identified in zebrafish Nos2b was found in two other clusters of this phylogenetic tree, namely the mammalian NOS3 and the cyprinidae NOS2. The overall score for a myristoylation prediction site, which was 0.7 for mammalian NOS3 sequences, was higher for fish NOS2 sequences (3 for carp and goldfish, 1 for zebrafish Nos2b sequences) and the probability of false positive prediction with the NMT software was lower for fish NOS2 than for mammalian NOS3. This phylogenetic tree suggests at least two rounds of duplications to explain the divergence of NOS isoforms in mammals. The first one could have led to constitutive (NOS1) and inducible (NOS2) NOS isoforms in vertebrates, whereas the second one would have given NOS1 and NOS3 sequences from a common nos1 ancestral gene in amphibian and mammals. Furthermore, in teleost such as zebrafish, an additional duplication event may have maintained two nos2 genes.

61

ksr2 and wsb2 respectively, suggesting that an ancestral genomic region containing nos, ksr and wsb genes did exist prior to duplication events (Fig. 3). Indeed, in the Ciona genome, we detected a nos gene and a wsb gene separated by 115 kb on chromosome K10q. In this urochordate specie, nos and wsb genes appear to be unique but no ksr ortholog could be identified in the sequence separating nos and wsb genes. In Xenopus, nos1 is linked to ksr2 and wsb2 on scaffold 340, and we found nos2 gene associated to nemo-like kinase gene (nlk), a gene present in nos2 syntenic region in human, murine and chicken genomes. In zebrafish, nos2b is located on chromosome15 and associated with ksr1, wsb1 and nlk in a 1038 kb fragment. On the other hand, nos2a is located on chromosome 5 and is not closely associated with any previously identified nos1 or nos2 syntenic gene. In zebrafish, nos1 is linked to ksr2 and wsb2 loci in a 500 kb fragment. In fugu, nos1 is associated with ksr2 and wsb2 on scaffold 4, whereas ksr1 and wsb1 are linked on scaffold 436 but no nos2 gene could be identified in this region. In human and mouse genomes, we did not identify any association of nos3 genes with paralogs of nos1 or nos2 syntenic gene. Nos3 gene overlaps with the atg9B gene (also named nos3 antisense gene) and is closely associated with abcb8 and cdk5 genes located on human chromosome 7 and mouse chromosome 5, respectively. In Xenopus, this synteny is also observed on scaffold 1193, and the nos3 and atg9 genes overlap. In zebrafish, there is no nos sequence in the genomic region containing atg9 and orthologs of nos3 syntenic genes.

3.2. Synteny analysis of nos genes among vertebrates Syntenic relationships with nos genes were then evaluated by comparing the genomic locations of nos homologs in human, mouse, chicken, Xenopus, zebrafish and Ciona genomes. In human, mouse and chicken genomes, nos2 genes are linked with the kinase suppressor of ras 1 (ksr1) gene and the WD repeat and SOCS boxcontaining protein 1(wsb1) gene. Interestingly, the nos1 gene is linked in these species to the respective paralogous genes of ksr1 and wsb1,

Fig. 2. Neighbor-joining phylogenetic tree of NOS protein sequences. Numbers at branch nodes represent bootstrap values. Asterisk indicates the sequence with predicted N-myristoylation site.

62

S. Lepiller et al. / Gene 445 (2009) 58–65

Fig. 3. Identification of conserved syntenies containing nos2 and nos1 genes in vertebrate species including human, mouse, chicken, Xenopus, zebrafish and Ciona. ksr: kinase suppressor of ras; wsb: WD repeat and SOCS box-containing protein; nlk : nemo-like kinase.

3.3. Comparison of nos2a and nos2b expression during development

3.5. Induction of nos2 genes expression by pro-inflammatory molecules

The temporal expression pattern of nos2a was studied by RT-PCR in embryos and larvae between 1 and 120 hpf and compared to nos2b expression. Both, nos2a and nos2b genes are maternally and zygotically expressed. Maternal expression is low in both cases (Fig. 4). The expression of nos2a mRNA is very low from 6 to 72 hpf, then increases from 96 hpf to be maintained at 120 hpf. The mRNA level of nos2b is detected as early as 6 hpf and is maintained throughout embryonic and larval development. Nos2b is expressed specifically in tissues surrounding the mouth in zebrafish larvae as evidenced by WISH experiments performed by Poon et al. (2008). We investigated the expression of nos2a in 48, 72 and 96 hpf old zebrafish larvae. In Fig. 5, we observed a specific labeling of the larvae with the antisense probe (Fig. 5B) compared to the absence of specific signal obtained with the sense riboprobe (Fig. 5A). Nos2a RNA a was mainly detected in the whole head, the eyes, and at 96 hpf a strong signal is observed in the gut region.

To investigate inducibility of nos2 zebrafish genes in vitro, we treated zebrafish cell lines with pro-inflammatory molecules. Exposure of zebrafish ZFL hepatocytes and ZF4 fibroblasts to either LPS or PolyI:C for 8 h induced expression of nos2a mRNA with the strongest induction observed in ZFL exposed to PolyI:C for 8 h (Fig. 7A and B). This induction was transient as nos2a mRNA level had returned to its basal level at 20 h. In contrast, nos2b was constitutively expressed in these cell lines and remained unchanged upon treatment with LPS or PolyI:C. The transient expression of nos2a RNA detected at 8 h postinduction of ZFL cells with polyI:C correlated with an increase of NO production by the cells. Indeed, we can observed by Griess assay an accumulation of NO metabolites (NOx) in supernatants of polyI:C treated ZFL cell lines 1 day after the stimulation (1.80 μM NOx with 50 μg/mL polyI:C, compared to 0.08 μM with 50 μg/mL LPS and 0.16 μM in control) and 2 days after stimulation (3.04 μM NOx with 50 μg/mL polyI:C, compared to 0.25 μM with 50 μg/mL LPS and 0.12 μM in control). We then investigated the inducibility of nos2 zebrafish genes in vivo. An increase in nos2a and, to a lesser extent, nos2b gene transcripts was identified at the site of injection 6 h after having challenged adult zebrafish with S. typhimurium LPS (Fig. 7C). The parallel induction of ifn1 interferon gene expression was used as a positive control (Altmann et al., 2003). Nos2a inducibility was also tested on the heart of adult zebrafish challenged with LPS or polyI:C. Adult zebrafishes were injected by ventral ip with PBS (50 μL), polyI:C (5 μg in 50 μL PBS) or LPS (5 μg in 50 μL PBS) and 6 h post-injection hearts were isolated from non injected and injected individuals. We have not observed in non injected fish any nos2a signal in the heart, but in injected fish a basal expression of nos2a is observed which correlated with the presence of ifn1 expression. In poly I:C injected fish a dramatic increase in nos2a expression is observed. So the expression of nos2a is inducible in fish heart.

3.4. Basal expression of the nos2 genes in adult zebrafish RT-PCR analysis of the expression of nos2a and nos2b transcripts in adult zebrafish organs indicated that nos2b transcripts were equally present in all tissues whereas nos2a transcripts showed a more restricted pattern of expression with the highest expression in spleen, muscle, skin, gut and ovary, a lower expression in kidney, gills, brain, eye, liver and testis and a lack of expression in the heart (Fig. 6).

3.6. Induction of nos2 genes expression by injury stress Fig. 4. Temporal expression pattern of nos2a and nos2b during zebrafish development. Analysis was performed by reverse transcriptase-polymerase chain reaction (RT-PCR) using embryos and larvae at indicated stages (hpf: hours post-fertilization). β-actin was used as a housekeeper gene control.

The transection of the tail created an inflammatory and regenerating site, where NO producing cells were detected by labeling with the NO specific DAF-FM DA probe, as previously described (Lepiller

S. Lepiller et al. / Gene 445 (2009) 58–65

63

Fig. 5. Spatial expression pattern of nos2a analyzed by WISH on 48, 72 and 96 hpf zebrafish. Specificity of the nos2a antisense probe (B colunn) was verified by using nos2a sense probe (A colunn).

et al., 2007). Tail section experiments were performed on 4 dpf old larvae and 10 h later, the expression of both nos2 isoforms was checked by WISH. A dramatic increase in nos2a and nos2b expressions is observed at the regeneration site and in skin cells (Fig. 8). Moreover, a specific nos2b signal is evidenced in neuromast cells of the lateral line. These data emphasized the inducible feature of both nos2a and nos2b zebrafish isoforms in stressful conditions. 4. Discussion 4.1. Zebrafish genome encodes two nos2 genes The present study demonstrates that, in addition to the ortholog of mammalian nos1 gene (Holmqvist et al., 2000; Poon et al., 2003), zebrafish genome encodes two nos2 genes. Both exhibit higher homologies with vertebrate nos2 than with constitutively expressed nos genes. This is a rare example of the presence of two complete and different nos2 genes in a vertebrate genome. The presence of duplicate genes is a well known phenomena in zebrafish (Amores et al., 1998; Nornes et al., 1998; Postlethwait et al., 1998; Lister et al., 2001; Bollig et al., 2006; Ishikawa et al., 2007; Rodriguez-Martin et al., 2007)

and can be explained by whole genome duplication that occurred during evolution of the ray-finned lineage (Volff, 2005). Phylogenetic analysis identified bony fish nos2 genes as a branch that was distinct from that leading from catshark to mammalian nos2, suggesting that a more ancient duplication event took place before chondrichthyan/ teleostome split, with one duplicate lost in each branch (RobinsonRechavi et al., 2004). Another explanation could be that an ancestral nos2 evolved faster in teleosts than in other vertebrates. Three different events were reported to occur after gene duplication. In most cases, one member of the pair evolves as a pseudogene through degenerative mutations and/or is eliminated because of its dispensability (nonfunctionalization). Alternatively, duplicate genes are preserved and remained functional. In this case, one copy can acquire a mutation conferring a new and beneficial function (neofunctionalization) whereas the other retains the original function. Lastly, duplicate genes can be preserved by subfunctionalization, whereby both copies dispatch the multiple functions of the ancestral gene between them (Lynch and Force, 2000). This mechanism of preservation of duplicate genes by complementary loss of subfunction is particularly associated with developmental genes and the evolution of many gene duplicates is consistent with this subfunctionalization model in zebrafish (Nornes et al., 1998; Force et al., 1999). 4.2. Divergence of the two zebrafish nos2 genes

Fig. 6. Expression of nos2a and nos2b genes in different zebrafish adult tissues. Total RNA was extracted from tissues isolated from 6 month old zebrafish for RT-PCR using nos2a and nos2b specific primers. Primers for β-actin gene were used as positive control.

Expression and synteny studies of zebrafish nos2 gene suggest that one paralog was maintained to perform the functions of the unique nos2 gene of higher vertebrates whereas the other seems to have gained new functions, as suggested by its N-myristoylation site and its constitutive and specific expression in development. The inducible nos2a is expressed in low amounts during embryonic development mainly in the head and increases significantly at 96 hpf, especially in the gut region. Nos2a is expressed in many adult organs, in particular in immune tissues such as spleen and major entries for pathogens such as gills, as well as skin and gut which was already

64

S. Lepiller et al. / Gene 445 (2009) 58–65

Fig. 7. Induction of nos2a and nos2b gene expression. ZFL (A) or ZF4 (B) cell lines remained unstimulated (c) or were stimulated with LPS or PolyI:C for 8 or 20 h. Adult zebrafish (C) were challenged with LPS (5 μg in 50 μL) or PBS (as a negative control). Total RNA from cells or site of injection was reversed-transcribed and amplified to detect nos2a, nos2b, ifn1 and the β-actin control.

reported in trout (Campos-Perez et al., 2000). In mammals, nos2 is also expressed constitutively in various cell types and tissues such as retina and skeletal muscles (Park et al., 1996; Punkt et al., 2002). Zebrafish nos2a expression is inducible in vitro and in vivo by proinflammatory molecules. Recently, induction of the zebrafish nos2 gene encoding the Nos2a isoform, was reported by activation of adult kidney cells with extracellular products of a bacterial pathogen Aeromonas hydrophila (Rodriguez et al., 2008) and this induction correlated with the release of nitrogen (NO) reactive free radicals. These data suggested that zebrafish Nos2 generates NO which behaves as a non-specific defense product as reported in other vertebrates. So among the two nos2 described in zebrafish, induction analyses indicate that nos2a behaves as a classical inducible NOS isoform whereas syntenic analyses show that nos2b gene is an ortholog of other nos2 vertebrate genes, localized in the expected genomic region. In contrast to the nos2a gene, the nos2b gene is constitutively expressed in specific tissues surrounding the oral cavity along development and in all zebrafish adult tissues analyzed. Its N-terminal myristoylation sequence suggests functional homologies with mammalian NOS3. A NOS3 was showed to be detected in various fishes and recently in lungfish, based on immunostaining data, (Fritsche et al., 2000; Amelio et al., 2006). Purification and N-terminal sequencing of the immunorevealed fish protein(s) should be interesting in order to characterize clearly the corresponding isoform. It could correspond to Nos1, phylogenetic nearer to mammal NOS3 sequences, or Nos2b which shared, as we show it, properties with NOS3 isoforms, since no nos3 nucleotidic sequence is related in fish databases. Accordingly, we did not find any nos3 gene sequence in Fugu, Medaka and Tetraodon

genome databases. It was suggested that the endothelial enzyme appeared in reptiles (Donald and Broughton, 2005) but a nos3 annotated sequence was identified in the Xenopus genome, which is clearly distinct from the nos1 and nos2 genes present in this species. It remains to be determined whether Xenopus NOS3 contains a myristoylation site. In mammals, myristoylation and palmitoylation target NOS3 to plasma membrane microdomains identified as caveolae, which may facilitate interaction with other proteins and with lipids to initiate signaling pathways (Dudzinski et al., 2006). The Nmyristoylation consensus site identified in Nos2b suggests specific functions for this early and constitutively expressed enzyme. Additional similarities between mammalian nos3 and zebrafish nos2b genes include their expression in adult heart and their induction upon exposure to LPS (Forstermann et al.,1998). Thus, zebrafish Nos2b could exert some of the functions of mammalian NOS3. Moreover, a strong induction of nos2b can be detected after tail transection, a property shared with nos2a gene, suggesting that their respective promoters contain similar regulatory regions. The evidenced expression of nos2b RNA in neuromasts cells, after tail section, illustrates the fact that this gene can be regulated in various cell types. Nos expression in these cells could be related to stress conditions and wound repair. 4.3. Variability in the number of nos genes in deuterostomes Whereas a nos2 homolog could not be found in the genomes of fugu, tetraodon, stickleback and medaka species, one nos2 gene was described in catshark, trout and carp, two nos2 forms were described in goldfish and two nos2 genes in zebrafish. Conversely, nos1 can be

Fig. 8. Induction of nos2a and nos2b gene expression after tail transection. WISH experiments are performed with nos2a (A, B) and nos2b (C, D) antisense riboprobes on 4 dpf control larvae (A, C) or on larvae on which tail transection was performed 10 h previously (B, D).

S. Lepiller et al. / Gene 445 (2009) 58–65

found in all vertebrates. Our analysis suggests that a nos ancestral gene was duplicated and gave rise to nos1 and nos2 loci. This ancestral gene could be sufficient to regulate many of the vital nitric oxide function in urochordate and in fishes whose one duplicated gene was secondarily lost. Two partial sequences of nos cDNA were cloned in the echinoderm sea urchin (Arbacia punctulata) and could represent early homologs of the NOS1 and NOS2 isoforms (Cox et al., 2001), indicating that NOS isoform diversity may exist early in deuterostomes. Thus, the number of nos loci can vary between species, especially in the heterogeneous phylogenetic group of fishes, since NOS isoforms could exert redundant functions as observed by studying nos knock-out consequences in mice (Mashimo and Goyal, 1999; Tsutsui et al., 2006). In conclusion, the zebrafish genome encodes two nos2 genes. Based on a synteny study, we attribute an orthologous position of zebrafish nos2b gene with mammalian nos2, since nos2a is located in another chromosome environment. Nos2a behaves as the mammalian NOS2 while Nos2b is more widely expressed in adult, possesses an N-terminal myristoylation consensus sequence and is detected in adult heart, which suggests functions similar to that of mammalian NOS3. These studies suggest an original evolutionary scenario giving rise to the NOS1 and NOS2 isoforms in vertebrates. Acknowledgments We thank Hervé Bouhin (UMR CNRS 5548, Dijon) and Thibault de Malliard (Master studient) for their help in phylogenetic analysis, Muriel Tauzin and Philippe Herbomel (Institut Pasteur, Paris) and Patrick Blader (UMR CNRS 5547, Toulouse) for their advices in wholemount in situ hybridization. We are grateful to Michael Schorpp (Max Planck Institut, Freiburg) for being so generous with his time and knowledge, welcoming V.L. to learn zebrafish methods in his lab. This work was supported by grants to J.C. from the Ligue Contre le Cancer Côte d'Or, Nièvre, Saône et Loire, the Association pour la Recherche contre le Cancer (ARC) and the Conseil Régional de Bourgogne. References Alderton, W.K., Cooper, C.E., Knowles, R.G., 2001. Nitric oxide synthases: structure, function and inhibition. Biochem. J. 357, 593–615. Altmann, S.M., Mellon, M.T., Distel, D.L., Kim, C.H., 2003. Molecular and functional analysis of an interferon gene from the zebrafish, Danio rerio. J. Virol. 77, 1992–2002. Amelio, D., Garofalo, F., Pellegrino, D., Giordano, F., Tota, B., Cerra, M.C., 2006. Cardiac expression and distribution of nitric oxide synthases in the ventricle of the coldadapted Antarctic teleosts, the hemoglobinless Chionodraco hamatus and the redblooded Trematomus bernacchii. Nitric Oxide 15, 190–198. Amores, A., et al., 1998. Zebrafish hox clusters and vertebrate genome evolution. Science 282, 1711–1714. Bollig, F., et al., 2006. Identification and comparative expression analysis of a second wt1 gene in zebrafish. Dev. Dyn. 235, 554–561. Bridle, A.R., Morrison, R.N., Nowak, B.F., 2006. The expression of immune-regulatory genes in rainbow trout, Oncorhynchus mykiss, during amoebic gill disease (AGD). Fish Shellfish Immunol. 20, 346–364. Campos-Perez, J.J., Ward, M., Grabowski, P.S., Ellis, A.E., Secombes, C.J., 2000. The gills are an important site of iNOS expression in rainbow trout Oncorhynchus mykiss after challenge with the gram-positive pathogen Renibacterium salmoninarum. Immunology 99, 153–161. Chenna, R., et al., 2003. Multiple sequence alignment with the Clustal series of programs. Nucleic Acids Res. 31, 3497–3500. Cox, R.L., Mariano, T., Heck, D.E., Laskin, J.D., Stegeman, J.J., 2001. Nitric oxide synthase sequences in the marine fish Stenotomus chrysops and the sea urchin Arbacia punctulata, and phylogenetic analysis of nitric oxide synthase calmodulin-binding domains. Comp. Biochem. Physiol. B, Biochem. Mol. Biol. 130, 479–491. Donald, J.A., Broughton, B.R., 2005. Nitric oxide control of lower vertebrate blood vessels by vasomotor nerves. Comp. Biochem. Physiol., A Mol. Integr. Physiol. 142, 188–197. Dudzinski, D.M., Igarashi, J., Greif, D., Michel, T., 2006. The regulation and pharmacology of endothelial nitric oxide synthase. Annu. Rev. Pharmacol. Toxicol. 46, 235–276.

65

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, 1531–1545. Forstermann, U., Boissel, J.P., Kleinert, H., 1998. Expressional control of the ‘constitutive’ isoforms of nitric oxide synthase (NOS I and NOS III). FASEB J. 12, 773–790. Fritsche, R., Schwerte, T., Pelster, B., 2000. Nitric oxide and vascular reactivity in developing zebrafish, Danio rerio. Am. J. Physiol. Regul. Integr. Comp. Physiol. 279, R2200–R2207. Gonzalez, S.F., Buchmann, K., Nielsen, M.E., 2007. Real-time gene expression analysis in carp (Cyprinus carpio L.) skin: inflammatory responses caused by the ectoparasite Ichthyophthirius multifiliis. Fish Shellfish Immunol. 22, 641–650. Holmqvist, B., et al., 2000. Identification and distribution of nitric oxide synthase in the brain of adult zebrafish. Neurosci. Lett. 292, 119–122. Ishikawa, T.O., Griffin, K.J., Banerjee, U., Herschman, H.R., 2007. The zebrafish genome contains two inducible, functional cyclooxygenase-2 genes. Biochem. Biophys. Res. Commun. 352, 181–187. Jeannin, J.-F., Leon, L., Cortier, M., Sassi, N., Paul, C, Bettaieb, A., 2008. Nitric oxide-induced resistance or sensitization to death in tumor cells. Nitric Oxide 19 (2), 158–163. Kimmel, C.B., Ballard, W.W., Kimmel, S.R., Ullmann, B., Schilling, T.F., 1995. Stages of embryonic development of the zebrafish. Dev. Dyn. 203, 253–310. Kroncke, K.D., Fehsel, K., Kolb-Bachofen, V., 1997. Nitric oxide: cytotoxicity versus cytoprotection—how, why, when, and where? Nitric Oxide 1, 107–120. Kroncke, K.D., Suschek, C.V., Kolb-Bachofen, V., 2000. Implications of inducible nitric oxide synthase expression and enzyme activity. Antioxid. Redox Signal. 2, 585–605. Laing, K.J., Grabowski, P.S., Belosevic, M., Secombes, C.J., 1996. A partial sequence for nitric oxide synthase from a goldfish (Carassius auratus) macrophage cell line. Immunol. Cell Biol. 74, 374–379. Lepiller, S., Laurens, V., Bouchot, A., Herbomel, P., Solary, E., Chluba, J., 2007. Imaging of nitric oxide in a living vertebrate using a diaminofluorescein probe. Free Radic. Biol. Med. 43, 619–627. Lin, A.W., Chang, C.C., McCormick, C.C., 1996. Molecular cloning and expression of an avian macrophage nitric-oxide synthase cDNA and the analysis of the genomic 5′flanking region. J. Biol. Chem. 271, 11911–11919. Lister, J.A., Close, J., Raible, D.W., 2001. Duplicate mitf genes in zebrafish: complementary expression and conservation of melanogenic potential. Dev. Biol. 237, 333–344. Lynch, M., Force, A., 2000. The probability of duplicate gene preservation by subfunctionalization. Genetics 154, 459–473. Mashimo, H., Goyal, R.K., 1999. Lessons from genetically engineered animal models. IV. Nitric oxide synthase gene knockout mice. Am. J. Physiol. 277, G745–G750. Nornes, S., et al., 1998. Zebrafish contains two pax6 genes involved in eye development. Mech. Dev. 77, 185–196. Park, C.S., Park, R., Krishna, G., 1996. Constitutive expression and structural diversity of inducible isoform of nitric oxide synthase in human tissues. Life Sci. 59, 219–225. Poon, K.L., Richardson, M., Korzh, V., 2008. Expression of zebrafish nos2b surrounds oral cavity. Dev. Dyn. 237, 1662–1667. Poon, K.L., Richardson, M., Lam, C.S., Khoo, H.E., Korzh, V., 2003. Expression pattern of neuronal nitric oxide synthase in embryonic zebrafish. Gene Expression Patterns 3, 463–466. Postlethwait, J.H., et al., 1998. Vertebrate genome evolution and the zebrafish gene map. Nat. Genet. 18, 345–349. Punkt, K., Naupert, A., Wellner, M., Asmussen, G., Schmidt, C., Buchwalow, I.B., 2002. Nitric oxide synthase II in rat skeletal muscles. Histochem. Cell Biol. 118, 371–379. Reddick, J.I., Goostrey, A., Secombes, C.J., 2006. Cloning of iNOS in the small spotted catshark (Scyliorhinus canicula). Dev. Comp. Immunol. 30, 1009–1022. Robinson-Rechavi, M., Boussau, B., Laudet, V., 2004. Phylogenetic dating and characterization of gene duplications in vertebrates: the cartilaginous fish reference. Mol. Biol. Evol. 21, 580–586. Rodriguez, I., Novoa, B, Figueras, A., 2008. Immune response of zebrafish (Danio rerio) against a newly isolated bacterial pathogen Aeromonas hydrophila. Fish Shellfish Immunol. 25 (3), 239–249. Rodriguez-Martin, I., et al., 2007. Characterization of two duplicate zebrafish Cb2-like cannabinoid receptors. Gene 389, 36–44. Saeij, J.P., Stet, R.J., Groeneveld, A., Verburg-van Kemenade, L.B., van Muiswinkel, W.B., Wiegertjes, G.F., 2000. Molecular and functional characterization of a fish inducible-type nitric oxide synthase. Immunogenetics 51, 339–346. Saitou, N., Nei, M., 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4, 406–425. Shin, D.H., et al., 2000. Immunocytochemical localization of neuronal and inducible nitric oxide synthase in the retina of zebrafish, Brachydanio rerio. Neurosci. Lett. 292, 220–222. Tsutsui, M., Shimokawa, H., Morishita, T., Nakashima, Y., Yanagihara, N., 2006. Development of genetically engineered mice lacking all three nitric oxide synthases. J. Pharmacol. Sci. 102, 147–154. Volff, J.N., 2005. Genome evolution and biodiversity in teleost fish. Heredity 94, 280–294. Wang, T., Ward, M., Grabowski, P., Secombes, C.J., 2001. Molecular cloning, gene organization and expression of rainbow trout (Oncorhynchus mykiss) inducible nitric oxide synthase (iNOS) gene. Biochem. J. 358, 747–755. Westerfield, M., 2000. The Zebrafish Book. A Guide for the Laboratory Use of Zebrafish (Danio rerio). University of Oregon, Eugene, OR.