Search

Cluster of differentiation 28 (CD28) is a co-stimulatory receptor found on the surface of T cells. Takifugu obscurus is a kind of anadromous fish species. In this study, the full-length sequence of To-CD28 was obtained, including a 672-bp open reading frame that encodes a peptide chain of 223 amino acids. The phylogenetic analysis showed that To-CD28 is similar to the CD28 protein in Takifugu rubripes. The total hematocyte count distinctly decreased after TBT-Cl exposure, showing the adverse effects of TBT-Cl invasion and self-adjusting ability upon To-CD28 accumulation. The production of reactive oxygen species increased, demonstrating the oxidation resistance of T. obscurus when exposed to TBT-Cl. The tissue expression patterns indicated To-CD28 is a widely distributed receptor in T. obscurus. Its high expression in the liver and gill suggests that To-CD28 could be potentially functioned in TBT-Cl toxic process. The mRNA levels of To-CD28 and relative genes in the TLR-MyD88 signal pathway were significantly up-regulated under TBT-Cl exposure. The immunohistochemistry also showed that the To-CD28 protein signal was enhanced under TBT-Cl exposure, which proved that the positive protection of To-CD28 for maintaining homeostasis. Our study indicated that To-CD28 might participate in the toxicity mechanism upon TBT-Cl exposure and regulate homeostasis stability of T. obscurus.

Fish skeletal muscle is comprised of fast muscle (FM) and slow muscle (SM), which constitutes 60% of total the body mass. Fish skeletal muscle can affect fish swimming activity, which is important for aquaculture due to its growth-potentiating effects. DNA methylation can influence gene expression level. We previously identified multiple differentially expressed genes (DEGs) between FM and SM in Takifugu rubripes. However, it is unknown if the expression levels of these DEGs are influenced by DNA methylation. In the present study, we used DNA methylation sequencing to study the DNA methylation profiles of FM and SM in T. rubripes. SM had higher overall methylation levels than FM. A total of 8479 differentially methylated genes (DMGs) and 3407 DMGs containing differentially methylated regions (DMRs) in the promoter regions between FM and SM were identified. After enrichment analysis, we found functionally relevant DMGs between FM and SM, including Kapca, Plcd3a, Plcd1, Pi3k, Tsp4b and Pgfrb in the hedgehog signaling pathway and phosphatidylinositol (PI)-related pathways. Due to the different methylation levels of these genes between FM and SM, the expression levels of Kapca, Plcd3a, Plcd1, Pi3k, and Tsp4b were higher in FM and Pgfrb was higher in SM. There were differences in the hedgehog signaling pathway and PI-related pathways between FM and SM. In SM, the cytokine-cytokine receptor interaction promoted focal adhesion, while ECM-receptor interactions promoted focal adhesion in FM. These results provide information regarding the difference between FM and SM in T. rubripes.

Carbonic anhydrase VI (CA VI) has been characterized as a secretory isozyme in mammals. Our present study confirmed the occurrence of CA VI in pufferfish ( Takifugu rubripes ). In this study, genomic sequence information for the CA VI of pufferfish was used for molecular cloning. We cloned a 1821 bp cDNA sequence, which consisted of a complete coding sequence of 1623 bp and a deduced amino acid sequence of 540 amino acids from the open reading frame. A BLAST search indicated that this protein exhibits 53%, 79%, and 67% identity with human, tilapia, and gar CA VI, respectively. It also shows 63%–77% identity with other fish CA VI-like sequences (zebrafish, Asian arowana, salmon, and large yellow croaker). Moreover, alignment of two or more sequences revealed that the protein sequence of pufferfish CA VI has 34%–37% identity with mammalian and fish CA II sequences. An NH 2 -terminal signal peptide of 18 amino acids in length was predicted in the pufferfish CA VI sequence. Three potential N-linked glycosylation sites and two cysteine residues (Cys-28 and Cys-209) that are likely to form one disulfide bond were present in pufferfish CA VI. In silico and phylogenetic analyses revealed that pufferfish CA VI is an extracellular secretory protein. Active site analysis indicated that this protein is a low-activity CA isozymes due to a characteristic Val/Ile substitution at position 207. Homology modeling of puffer CA VI was performed using the crystal structure of human carbonic anhydrase XIV as a template structure, based on high similarity. Reverse transcription-polymerase chain reaction (PCR), quantitative PCR and in situ hybridization results revealed that, the pufferfish CA VI is highly expressed in liver tissue.

The fish insulin/insulin-like growth factor (IGF) pathway has weak control over carbohydrate metabolism. To understand the molecular basis for the metabolic diversity, we characterized the forkhead box transcription factor O1A (FoxO1A), a downstream target of the insulin/IGF pathway, in torafugu Takifugu rubripes. The cloned torafugu FoxO1A cDNA contained all conserved features critical for its transcriptional activity and a unique unspliced intron encoding a poly-glutamine stretch. Torafugu FoxO1A showed the IGF-dependent nuclear exclusion and in vitro binding to the well-conserved FoxO1 binding site, DAF-16 binding element (DBE), but failed to bind to the insulin-responsive element by which mammalian FoxO1 mediates insulin effects. The subsequent in silico genomic screening provided a list of 587 potential torafugu FoxO1A target genes containing the DBE. Some carbohydrate metabolic genes regulated by FoxO1 in mammals were not included in the list. We further identified about 250 potential fish FoxO1 target genes by integrating results of the DBE screening against fish metagenome that contained 262 species. Neuronal processes appeared to be the common major function of fish FoxO1, although further annotation of the potential target genes is required. These results provide a part of the molecular basis underlying the weak association between the insulin/IGF pathway and carbohydrate metabolism in fish.

Carbonic anhydrase (CA) is a key element for maintaining acid base balance in fish. In our present experiment, novel CA isozymes were identified from the pear puffer (Takifugu vermicularis). Based on the high homology of two predicted CA sequences of the tiger puffer (Takifugu rubripes), a 1715bp novel cDNA was obtained from T. vermicularis. The open reading frame showed a complete coding sequence of 552bp with a deduced peptide sequence of 183 amino acids that exhibited highest (97%) identity with pufferfish putative CA III and CA IV-like sequences. In addition, this translated protein sequence showed 36-37% identity with zebrafish CA IV-like, CA XVa, CA XVb, and CA XVc proteins. Phylogenetic analysis revealed that the pufferfish novel protein (pCAn) was a membrane-bound CA protein. Alignment of multiple CA sequences illustrated that most of the putative active site residues of the pCAn isozyme were situated at highly conserved regions of the CA sequences. Examination of motif distribution suggested that the pCAn isozyme was very similar to the puffer predicted CA IV-like isozyme. Reverse transcription-polymerase chain reaction (PCR) analysis showed highly differential expression in the brain, gills, kidney, and muscle, whereas CA mRNA expression was almost absent in heart, liver, and intestine. Quantitative PCR expression of CA mRNA abundance suggested several-fold higher expression of pCAn isozymes in the gills compared to other tissues tested. Our results suggest that the pCAn isozyme might be related to CA IV-like isozymes. Further functional studies are needed to investigate the function of the pCAn isozyme in T. vermicularis.

Neuregulin receptor degradation protein-1 (Nrdp1) was recently identified in humans as an important immune factor responding to the challenge of virus, LPS or cytokine. Its role in fish immune defense and whether it is involved in anti-parasite immunity have not been proven yet. In this report, the full-length cDNA sequence and genomic structure of Nrdp1 in the large yellow croaker Larimichthys crocea (LcNrdp1) were identified and characterized. The full-length cDNA of LcNrdp1 was 1248bp, including a 5' untranslated region (UTR) of 32bp, a 3' UTR of 259bp and an open reading frame (ORF) of 937bp, encoding a polypeptide of 318 amino acid residues. The full-length genomic DNA sequence of LcNrdp1 was composed of 2635 nucleotides, including four exons and three introns. The putative LcNrdp1 protein had no signal peptide sequence and contained a characteristic Nrdp1 consensus motif C3HC3D ring finger and a Coiled-coil domain. Phylogenetic analysis showed that Nrdp1 in fish was closer with that in other vertebrates (79%-90% amino acid identity) than in invertebrates and bacteria (27%-65%). In fishes, Nrdp1 in large yellow croaker was closer with that in Takifugu rubripes. The expression profile showed that LcNrdp1 was constitutively expressed in all tested tissues, especially highly expressed in brain, muscle and kidney. Post-infection (PI) with Cryptocaryon irritans, an increased expression of LcNrdp1 was induced in infection sites (skin and gill), whereas in immune organs, the expression of LcNrdp1 was up-regulated in spleen (except the 1st d and 10th d PI) but suppressed in head kidney. These results suggested that LcNrdp1 might play an important immune role in the finfish L. crocea in the defense against the parasite C. irritans.

MicroRNAs are indispensable players in the regulation of a broad range of biological processes. Here, we report the first deep sequencing of the whitespotted bamboo shark (Chiloscyllium plagiosum) liver. We mapped 91 miRNAs in the Callorhinchus milii genome that have previously been described in the Danio rerio, Fugu rubripes, Oryzias latipes, Xenopus laevis, Xenopus tropicalis, Homo sapiens, and Mus musculus. In addition, 156 new putative candidate (PC) C. plagiosum miRNAs were identified. From these 247 miRNAs, 39 miRNA clusters were identified, and the expression of these clustered miRNAs was observed to vary significantly. A total of 7 candidate miRNAs were selected for expression confirmation by stem-loop RT-PCR. This study resulted in the addition of a significant number of novel miRNA sequences to GenBank and laid the foundation for further understanding of the function of miRNAs in the regulation of C. plagiosum liver development.

In vertebrates, the development-dependent and tissue-specific expression of myosin heavy chain (MYH) genes (MYHs) contributes to the formation of diverged muscle fiber types. The expression patterns of developmentally regulated MYHs have been investigated in certain species of fish. However, the expression profiles of MYHs during torafugu Takifugu rubripes development, although extensively studied in adult tissues, have not been sufficiently studied, and also the expression orders of MYHs during development have remained unclear. In the present study, we comprehensively cloned four MYHs (MYH(M743-2), MYH(M86-2), MYH(M5) and MYH(M2126-1)) from embryos, and two MYHs (MYH(M2528-1) and MYH(M1034)) from larvae, and characterized their expression pattern in relation to developmental stages of torafugu by reverse transcription (RT)-PCR and in situ hybridization. The expression of MYHs from torafugu embryos and larvae appeared sequentially and varied largely in relation to the developmental stage-dependent and fibers-type-specific manners. The transcripts of MYH(M743-2) appeared first in embryos at 3 days post fertilization (dpf) and were localized in the epaxial and hypaxial domains of fast muscle fibers of larval myotome, whereas those of MYH(M5) and MYH(M86-2) in 3 dpf and 4 dpf, respectively, and both were localized in superficial slow and horizontal myoseptum regions. The expression of MYH(M1034) and MYH(M2126-1) was quite low and mostly undetectable. Different MYHs from torafugu embryos and larvae have also been found to be expressed differentially in pectoral fin and craniofacial muscles. Interestingly, the transcripts of MYH(M2528-1) first appeared at 6 dpf and were distinctly expressed at the dorsal and ventral extremes of larval myotome, suggesting its involvement in stratified hyperplasia. The novel involvement of MYH(M2528-1) in mosaic hyperplasia was further confirmed in juvenile torafugu, where the transcripts were expressed in fast fibers with small diameters as well as the inner part of superficial slow fibers.

Large-scale features of the spatial arrangement of protein-coding segments (PCS) are investigated by means of the inter-PCS spacers' size distributions, which have been found to follow power-laws. Linearity in double-logarithmic scale extends to several orders of magnitude in the genomes of organisms as disparate as mammals, insects and plants. This feature is also present in the most compact eukaryotic genomes and in half of the examined bacteria, despite their very limited non-coding space. We have tried to determine the sequence of events in the course of genomes' evolution which may account for the formation of the observed size distributions. The proposed mechanism essentially includes two types of events: (i) segmental duplications (and possibly paleopolyploidy), and (ii) the subsequent loss of most of the duplicated genes. It is shown by computer simulations that the formulated scenario generates power-law-like inter-PCS spacers' size distributions, which remain robust for a variety of parameter choices, even if insertion of external sequences, such as viruses or proliferating retroelements is included. Moreover, power-laws are preserved after most of the non-coding DNA has been removed, thus explaining the finding of this pattern in genomes as compact as that of Takifugu rubripes.

International audience; Vitellogenin (Vtg) derivatives are the main egg-yolk proteins in most oviparous animal species, and are, therefore, key players in reproduction and embryo development. Conserved synteny and phylogeny were used to identify a Vtg gene cluster (VGC) that had been evolutionarily conserved in most oviparous vertebrates, encompassing the three linked Vtgs on chicken (Gallus gallus) chromosome 8. Tandem arranged homologs to chicken VtgII and VtgIII were retrieved in similar locations in Xenopus (Xenopus tropicalis) and homologous transcribed inverted genes were found in medaka (Oryzias latipes), stickleback (Gasterosteus aculeatu), pufferfish (Takifugu rubripes), and Tetrahodon (Tetraodon nigroviridis), while zebrafish (Danio rerio) Vtg3 may represent a residual trace of VGC in this genome. Vtgs were not conserved in the paralogous chromosomal segment attributed to a whole-genome duplication event in the ancestor of teleosts, while tandem duplicated forms have survived the recent African clawed frog (Xenopus laevis) tetraploidization. Orthologs to chicken VtgI were found in similar locations in teleost fish, as well as in the platypus (Ornithorhynchus anatinus). Additional Vtg fragments found suggested that VGC had been conserved in this egg-laying mammal. A low ratio of nonsynonymous-to-synonymous substitution values and the paucity of pseudogene features suggest functional platypus Vtg products. Genomic identification of Vtgs, Apob, and Mtp in this genome, together with maximum likelihood and Bayesian inference phylogenetic analyses, support the existence of these three large lipid transfer protein superfamily members at the base of the mammalian lineage. In conclusion, the establishment of a VGC in the vertebrate lineage predates the divergence of ray-finned fish and tetrapods and the shift in reproductive and developmental strategy observed between prototherians and therians may be associated with its loss, as shown by its absence from the genomic resources currently available from therians. (c) 2008 Elsevier B.V. All rights reserved.

Several cannabinoid receptors have been detected in many organisms. The best known are CB1, mainly expressed in the central nervous system and CB2 which is almost exclusively expressed in the periphery. Here we report the molecular characterization of two duplicate CB2-like cannabinoid receptors from zebrafish (Danio rerio) (zebrafish Cb2a and zebrafish Cb2b). The amino acid sequences of these receptors present 56% identity with Takifugu rubripes CB2 sequence and 39% with human CB2 sequence and conserve some specific key residues for cannabinoid receptor function. Both duplicate receptors are expressed in peripheral tissues (gills, heart, intestine and muscle), immune tissue (spleen) and also in the central nervous system. Using in situ hybridization techniques zebrafish Cb2 mRNA expression was observed for the first time in the adenohypophysial cells of the rostral pars distalis and proximal pars distalis of the pituitary gland. Given the importance of the existence of duplication of genes in teleosts, the combined analysis of these two new cannabinoid receptors opens a new exciting door to investigate and understand cannabinoid function throughout evolution.

We have characterized the apolipoprotein multigene family of the pufferfish Takifugu rubripes. The pufferfish mainly contains 28-kDa, 27-kDa, and 14-kDa apolipoproteins in its plasma and was designated apo-28 kDa, apo-27 kDa, and apo-14 kDa, respectively. N-terminal amino acid sequencing revealed that pufferfish apo-28 kDa and apo-27 kDa have an identical amino acid sequence except an additional propeptide in the former; and both are homologues of apoA-I from other animals. The sequence of pufferfish apo-14 kDa is homologous to that of eel apo-14 kDa previously reported, both being apparently specific to fish. In silico screening, using the publicly available Fugu genome database confirmed the pufferfish apoA-I and apo-14 kDa genes. The database further contained the genes encoding four types of apoA-IV, one apoC-II and two types of apoE. Thus, pufferfish contains nine genes encoding apolipoprotein multigene family. Two apoA-IV and one apoE genes were tandemly arrayed and located on one scaffold. Thus two sets of these genes formed two gene clusters. The apoC-II and apo-14 kDa genes are also located on a single scaffold. apoA-I and apo-14 kDa gene transcripts were mainly expressed in liver and less abundantly in brain. The transcripts of the former gene were also observed in intestine. In contrast, the transcripts encoding four apoA-IVs, one apoC-II, and two apoEs were mainly expressed in intestine. These structural details of pufferfish apolipoproteins and tissue distribution of their gene transcripts provide a novel evidence for better understanding of evolutionary relationships of apolipoprotein multigene family.

Because of its unusual high degree of compaction and paucity of repetitive sequences, the genome of the smooth pufferfish Tetraodon nigroviridis is the subject of a well-advanced sequencing project. An astonishing diversity of transposable elements not found in the human and the mouse has been observed in the genome of T. nigroviridis. Due to the difficulty of assembling repeat-rich regions, the whole genome shotgun sequencing approach will probably fail to reveal the general organisation of this compact vertebrate genome. Therefore, in order to gain new insights into the global distribution pattern of repeated DNA in the genome of T. nigroviridis, we have reconstructed partial/complete repetitive sequences from data generated by the genome project and performed double-colour fluorescent in situ hybridization (FISH) analysis for representatives of three major categories of repeated sequences including two minisatellites (ms100 and ms104), two DNA transposons (Tol2 and Buffy1) and two non-long terminal repeat (LTR) retrotransposons (Rex3 and Babar). We show that DNA transposons and retroelements very frequently colocalize with minisatellites and mostly accumulate within heterochromatic regions. These results, which have not been reported so far for the fugu Takifugu rubripes, show that repeated elements are generally excluded from gene-rich regions in T. nigroviridis and underline the extreme degree of compartmentalization of this compact genome. The genome organization of the pufferfish is clearly different from that observed in humans, where repeated sequences make up an important fraction of euchromatic DNA, and is more similar to that observed in the fruit fly Drosophila melanogaster.

Beta-defensins (β-defensins) are small cationic amphiphilic peptides that are widely distributed in plants, insects, and vertebrates, and are important for their antimicrobial properties. In this study, the β-defensin (Onβ-defensin) gene of the Nile tilapia (Oreochromis niloticus) was cloned from spleen tissue. Onβ-defensin has a genomic DNA sequence of 674 bp and produces a cDNA of 454 bp. Sequence alignments showed that Onβ-defensin contains three exons and two introns. Sequence analysis of the cDNA identified an open reading frame of 201 bp, encoding 66 amino acids. Bioinformatic analysis showed that Onβ-defensin encodes a cytoplasmic protein molecule containing a signal peptide. The deduced amino acid sequence of this peptide contains six conserved cysteine residues and two conserved glycine residues, and shows 81.82% and 78.33% sequence similarities with β-defensin-1 of fugu (Takifugu rubripes) and rainbow trout (Oncorhynchus mykiss), respectively. Real-time quantitative PCR showed that the level of Onβ-defensin expression was highest in the skin (307.1-fold), followed by the spleen (77.3-fold), kidney (17.8-fold), and muscle (16.5-fold) compared to controls. By contrast, low levels of expression were found in the liver, heart, intestine, stomach, and gill (

Pax3 and Pax7 are the regulators and markers of muscle progenitors and satellite cells that contribute to the embryonic development and postembryonic growth of skeletal muscle in vertebrates, as well as to its repair and regeneration. However, information regarding them in vertebrate genome model, torafugu Takifugu rubripes, has remained unknown. Therefore, as an initial step, here we characterized Pax3 and Pax7 from torafugu and investigated their expression patterns during different developmental stages by RT-PCR. In silico analysis with the Fugu genome database (ver. 4.0) yielded two distinct genes each for Pax3 (Pax3a and Pax3b) and Pax7 (Pax7a and Pax7b). The 75th amino acid, glutamine (Gln75), from the N-terminus was replaced by proline in the paired box domain (PD) of Pax3a. One single cDNA clone encoding Pax3a had deletion of Gln75 in PD, suggesting the presence of alternatively spliced variants (Q+/Q-). This was further supported by identification of two adjacent alternative 3' splice acceptor sites which produce Pax3b Q+ (aagCAGGGA) and Q- (aagcagGGA) variants. Interestingly, torafugu Pax7a, but not Pax7b, had an insert encoding five amino acid residues (SGEAS) in a C-terminal region of PD in two out of three cDNA clones. Genomic analysis showed two alternate splice donor sites at exon 4 of Pax7a. In synteny analysis, torafugu Pax3a showed syntenic relationship with the corresponding regions in other teleosts only, whereas Pax3b and Pax7b showed high syntenic relationship with the corresponding regions of both mammals and other teleosts. RT-PCR revealed that expression of Pax3a and Pax3b transcripts was restricted to embryonic stages only, whereas those of Pax7a and Pax7b was continued to be expressed in larvae and importantly those of Pax7a were found in adult skeletal muscles. Therefore, Pax3 appears to be most important for primary myogenesis and Pax7 for secondary myogenesis and growth by hyperplasia in fish. In this regard, the transcripts of torafugu Pax3 and Pax7 genes might be used for further investigation as a marker for identification of muscle precursor cells during different phases of growth, and this ambiguity is the next target of our research.

Splice variants of the basic helix-loop-helix myoblast determination factor (myoD) have not been previously found in vertebrates. Here we report the identification and characterization of three alternative transcripts of a myoD paralogue from the tiger pufferfish (Takifugu rubripes). The T. rubripes myoD1 gene (TmyoD1) has 3 exons and 2 introns and it is present on scaffold 104, in a region of conserved synteny with zebrafish. The isoform TMyoD1-alpha is a putative protein of 281 residues that contains the basic, helix-loop-helix and helix III domains and shares 61%, 56%, 51%, 49% and 56% overall identity with zebrafish, Xenopus, mouse, human and chicken MyoD1, respectively. TMyoD1-beta arises from an alternative 3' splice site and differs from TMyoD1-alpha by a 26-residue insertion adjacent to helix III, which is one of the functional domains required for chromatin remodelling. The third alternative transcript, TmyoD1-gamma, retains intron I and has two premature termination codons far from the 3'-most exon-exon junction. TmyoD1-gamma is therefore likely to be degraded by nonsense-mediated decay, an important widespread post-transcriptional mechanism that regulates transcript levels. Analysis of gene expression by qPCR revealed that TmyoD1-alpha was the most abundant transcript in fast and slow myotomal muscle. TmyoD1-alpha expression was 2-fold higher in fast muscle of juvenile fish that were actively producing new myotubes compared to adult stages that had stopped recruiting fast muscle fibres. A similar expression pattern was observed for TmyoD1-alpha in slow muscle but the differences were not significant. Transcript levels of TmyoD1-gamma only varied significantly in fast muscle and were 5-fold higher in adult compared to juvenile stages. Significant differences in expression of TmyoD1 splice variants were also observed during embryonic development. The differential expression of three alternative transcripts of myoD1 in developing and adult myotomal muscle of T. rubripes supports the hypothesis that diversity generated by alternative splicing may be of functional significance in muscle development in this species.

Using gene cloning and in silico cloning, we analyzed the structures of hatching enzyme gene orthologs of vertebrates. Comparison led to a hypothesis that hatching enzyme genes of Japanese eel conserve an ancestral structure of the genes of fishes, amphibians, birds and mammals. However, the exon-intron structure of the genes was different from species to species in Teleostei: Japanese eel hatching enzyme genes were 9-exon-8-intron genes, and zebrafish genes were 5-exon-4-intron genes. In the present study, we further analyzed the gene structures of fishes belonging to Acanthopterygii. In the species of Teleostei we examined, diversification of hatching enzyme gene into two paralogous genes for HCE (high choriolytic enzyme) and LCE (low choriolytic enzyme) was found only in the acanthopterygian fishes such as medaka Oryzias latipes, Fundulus heteroclitus, Takifugu rubripes and Tetraodon nigroviridis. In addition, the HCE gene had no intron, while the LCE gene consisted of 8 exons and 7 introns. Phylogenetic analysis revealed that HCE and LCE genes were paralogous to each other, and diverged during the evolutionary lineage to Acanthopterygii. Analysis of gene synteny and cluster structure showed that the syntenic genes around the HCE and LCE genes were highly conserved between medaka and Teraodon, but such synteny was not found around the zebrafish hatching enzyme genes. We hypothesize that the zebrafish hatching enzyme genes were translocated from chromosome to chromosome, and lost some of their introns during evolution.

In this paper, we provide evidence for the body temperature effect on the formation of GC-rich isochores, by analysing genomic sequences from two puffer fishes living at different temperatures. The higher body temperature of Tetraodon nigroviridis compared to Takifugu rubripes (DeltaT approximately 15 degrees C) appears to be the cause of a higher compositional heterogeneity of the former due to the formation of GC-rich regions. Such an effect does not only concern large DNA segments but also coding sequences.

The tyrosinase gene family encompasses three members, tyrosinase, tyrosinase-related protein 1 (Tyrp1) and dopachrome tautomerase (Dct), which encode for proteins implicated in melanin synthesis. In human and mouse, genomic organization is known for all three genes, revealing common features of regulatory elements and of exon/intron structure. We have set out to identify the complete family from a more primitive vertebrate, the pufferfish Fugu (Takifugu rubripes), which is characterized by a compact genome. We had recently isolated and characterized the Fugu tyrosinase gene (Genesis 28 (2000) 99-105). We now report the isolation and characterization of the two other members of the family, Tyrp1 and Dct. Regulatory sequences from these genes function in mouse pigment cells and are able to mediate reporter gene expression. Our results demonstrate the existence of all three tyrosinase family members in teleosts and underline the evolutionary conservation of the pigmentary system.