The evolutionary relationship between the duplicated copies of the zebrafish fabp11 gene and the tetrapod FABP4, FABP5, FABP8 and FABP9 genes Santhosh Karanth 1 , Eileen M. Denovan-Wright 2 , Christine Thisse 3 , Bernard Thisse 3 and Jonathan M. Wright 1 1 Department of Biology, Dalhousie University, Halifax, Canada 2 Department of Pharmacology, Dalhousie University, Halifax, Canada 3 Department of Cell Biology, University of Virginia Health Sciences Center, Charlottesville, VA, USA The multigene family coding for vertebrate intra- cellular lipid-binding proteins (iLBPs) consists of the fatty acid-binding protein (FABP), cellular retinoic acid-binding protein (CRABP) and cellular retinol- binding protein (CRBP) genes. FABPs bind selectively to fatty acids, CRABPs bind to retinoic acid, and CRBPs bind to retinol [1]. For many iLBPs, the precise physiological function(s) is not completely understood or remains unknown. However, it is clear that iLBPs are involved in cellular uptake and intracel- lular transport of long-chain fatty acids, bile salts and retinoids, protection of cellular structures from the detergent effects of fatty acids by sequestering them until required in various metabolic processes, interaction Keywords Fabp11; retinal development; spinal cord; tandem gene duplication; whole genome duplication Correspondence J. M. Wright, Department of Biology, Dalhousie University, Halifax, NS, Canada B3H 4J1 Fax: +1 902 494 3736 Tel: +1 902 494 6468 E-mail: jmwright@dal.ca Website: http://www.dal.ca/biology2/ (Received 23 January 2008, revised 8 March 2008, accepted 9 April 2008) doi:10.1111/j.1742-4658.2008.06455.x We describe the structure of a fatty acid-binding protein 11 (fabp11b) gene and its tissue-specific expression in zebrafish. The 3.4 kb zebrafish fabp11b is the paralog of the previously described zebrafish fabp11a, with a deduced amino acid sequence for Fabp11B exhibiting 65% identity with that of Fabp11A. Whole mount in situ hybridization of a riboprobe to embryos and larvae showed that zebrafish fabp11b transcripts were restricted solely to the retina and were first detected at 24 h postfertilization. In situ hybrid- ization revealed fabp11b transcripts along the spinal cord in adult zebrafish. However, the highly sensitive RT-PCR assay detected fabp11b transcripts in the brain, heart, ovary and eye in adult tissues. By contrast, fabp11a transcripts had been previously detected in the liver, brain, heart, testis, muscle, ovary and skin of adult zebrafish. Using the LN54 radiation hybrid panel, we assigned zebrafish fabp11b to linkage group 16. Phylogenetic analysis and conserved gene synteny with tetrapod genes indicated that the emergence of two copies of fabp11 in the zebrafish genome may have resulted from a fish-specific whole genome duplication event. Furthermore, we propose that the FABP4–FABP5–FABP8–FABP9 (PERF15) gene cluster on a single chromosome in the tetrapod genome and the fabp11 genes in the zebrafish genome originated from a common ancestral gene, which, following their divergence, gave rise to the fabp11 genes of zebrafish, and the progenitor of the FABP4, FABP5, FABP8 and FABP9 genes in tetrapods after the separation of the fish and tetrapod lineages. Abbreviations CRABP, cellular retinoic acid-binding protein; CRBP, cellular retinol-binding protein; EST, expressed sequence tag; FABP, fatty acid-binding protein; hpf, hours postfertilization; iLBP, intracellular lipid-binding protein; LG, linkage group; mya, million years ago; WGD, whole genome duplication. FEBS Journal 275 (2008) 3031–3040 ª 2008 The Authors Journal compilation ª 2008 FEBS 3031 with other enzymes and transport systems, and the transcriptional regulation of specific genes [2–6]. By functioning in the transport and metabolism of retinol and retinoic acid, CRBPs and CRABPs may play an important role in development, growth and reproduc- tion, primarily by making retinoids available to recep- tors in the nucleus to regulate specific gene transcription [7]. Originally, FABPs and their genes were named on the basis of the tissue in which they were first isolated. Later, Hertzel & Bernlohr [8] pro- posed a different nomenclature, in which FABPs are numbered according to the temporal order of their identification (e.g. fabp1 and fabp2). We chose, for the sake of clarity, to use here the nomenclature adopted by Hertzel & Bernlohr [8]. Thus far, iLBPs have only been found in species from the animal kingdom, suggesting that a single ancestral iLBP gene emerged in animals after their divergence from plants and fungi [9]. The diversity of the iLBP multigene family is thought to have arisen through a series of gene duplication events followed by their sequence divergence [10]. Estimates of the earliest iLBP gene duplication vary between 930 and 1000 million years ago (mya) [9–11]. Schaap et al. [9] and Schleicher et al. [10] have suggested that 600–700 mya, before the invertebrate and vertebrate lineages split, the gene(s) that would give rise to the FABP1– FABP2–FABP6 clade had already diverged from the gene(s) that would give rise to the FABP4–FABP8 clade. CRBPs and CRABPs, which are absent in inver- tebrates, might have diverged from the FABP1 clade after the vertebrates and invertebrates split. To date, paralogs of 11 genes coding for FABPs have been described in vertebrates [12]. Zebrafish have attracted the attention of evolu- tionary molecular biologists partly because of the abundance of genetic and biological resources for this model organism for developmental studies, but partic- ularly for the whole genome duplication (WGD) event that occurred in ray-finned fishes some 250–400 mya [13], which has led to investigations on the genesis and fate of duplicated genes. Gene duplication has been proposed by Ohno [14] as a major evolutionary force in driving the increasing complexity of life. In addition to WGD, tandem duplication of individual genes by the process of unequal crossing-over during meiosis may also account for the increase in the number of genes in eukaryotes [15]. Vayda et al. [16] described a fabp gene from four Antarctic fishes, termed H ad -FABP, which they suggest is the ortholog of mammalian FABP4. In a previous communication [17], we described an fabp4 gene from zebrafish that showed greatest sequence identity to, and formed a clade in phylogenetic analyses with, the Antarctic fish H ad -FABP. Recently, however, Agulleiro et al. [12] recognized that the Fabps reported by Vayda et al. [16] and by us [17] constitute a new FABP clade that is probably restricted to fishes, and renamed the novel fish gene and protein fabp11 and Fabp11, respec- tively [12]. In this article, we report the charac- terization of a duplicated Fabp11 gene (fabp11b) from the zebrafish genome, the distribution of fabp11b mRNA transcripts in adult tissues and during embry- onic and larval development, and linkage group (LG) (chromosome) assignment of fabp11b by radiation hybrid mapping. On the basis of phylogenetic analysis and conserved gene syntenies of zebrafish fabp11a and fabp11b with other vertebrate FABP genes, we propose that the duplicated fabp11 genes in fishes and the FABP4–FABP5–FABP8–FABP9 gene cluster in tetra- pods arose from a single progenitor gene. Results and Discussion Identification of a duplicated fabp11 gene from zebrafish A paralogous gene to the previously described zebra- fish fabp11 (fabp4, but now referred to as fabp11a) [17] was identified from a blast search of the zebrafish genome sequence database at the Wellcome Trust Sanger Institute (version Zv6, http://www.ensembl.org/ Danio_rerio/index.html), using as query the GenBank sequence AY628221. The 3.4 kb duplicated fabp11 (hereafter referred to as fabp11b) consists of four exons and three introns (Fig. 1), a FABP gene organization seen for most vertebrate iLBP genes [1]. The four exons of fabp11b code for 24, 59, 34 and 17 amino acids, respectively (Fig. 1), identical to what is seen for zebrafish fabp11a [17]. For fabp11b, the splice junc- tions for intron 1 and intron 2 conform to the GT ⁄ AG intron ⁄ exon rule [18], whereas the splice junction for intron 3 is TA ⁄ AG. A putative binding site for a GATA transcription factor was identified at position )471 to )483 in the 5¢-upstream region of fabp11b (Fig. 1). Divine et al. [19] reported that GATA-4, GATA-5 and GATA-6 act cooperatively in activating FABP1 transcription in the murine small intestine. A putative TATA box is located at position )30 to )24. Distribution of fabp11b transcripts in the retina of developing zebrafish embryo and larvae To determine the spatial and temporal distribution of fabp11b transcripts during zebrafish embryonic and larval development, we conducted whole mount in situ Duplicated fabp11 genes of zebrafish S. Karanth et al. 3032 FEBS Journal 275 (2008) 3031–3040 ª 2008 The Authors Journal compilation ª 2008 FEBS hybridization to zebrafish embryos and larvae at different developmental stages (Fig. 2). fabp11b tran- scripts were first detected in the retina of developing embryos at 24 h postfertilization (hpf) in the pig- mented epithelium, starting in the proximal part of the retina. fabp11b transcripts spread across this layer of the retina by 30 hpf (Fig. 2B). A homogeneous distri- bution of fabp11b transcripts throughout the pig- mented epithelium was observed at 48 hpf (Fig. 2C). In contrast to fabp11b, at 24 hpf and 36 hpf the fabp11a transcripts were detected in the lens and diencephalon [17]. In 5-day-old larvae, the hybridiza- tion signal for fabp11b transcripts was still observed in the same layer of the retina (Fig. 2D). Sellner [20] reported a similar trend, whereby retinal FABP appears to be maximally expressed in the chicken ret- ina around the ninth day of embryonic development and declines at later stages. Embryonic development in zebrafish spans 3 days, whereas embryonic develop- ment in chicken extends over 20–21 days. Two other FABP transcripts, transcripts for fabp3 and fabp7b, have been detected in the developing retina of zebra- fish embryos [17,21]. It has been proposed that FABPs are involved in sequestering of fatty acids during retinal differentiation [20]. Tissue-specific distribution of fabp11b transcripts in adult zebrafish In situ hybridization was performed on sections of adult zebrafish to determine the tissue-specific distribu- tion of fabp11b transcripts. fabp11b transcripts were detected along the vertebra in the spinal cord of adult zebrafish (Fig. 3A shows the hybridization signal in a sagittal section, and Fig. 3B a transverse section). While describing the distribution of FABP8 expression in the rabbit spinal cord, Narayanan et al. [22] noted that the spinal cord has a high rate of fatty acid biosynthesis. RT-PCR detected fabp11b transcripts in total RNA extracted from the brain, heart, ovary and Fig. 1. The sequence of zebrafish fabp11b and its 5¢-upstream promoter region. Exons are shown in upper-case letters, with the coding sequences of each exon underlined and the deduced amino acid sequence indi- cated below. The stop codon is indicated by the diamond symbol. Only partial nucleotide sequences are shown for introns 2 and 3, where the dotted lines indicate interruption in the sequence. +1 indicates the transcrip- tion start site. A putative polyadenylation signal, AATAAA, is highlighted in bold and underlined. A putative TATA box and a site for GATA-binding factor are highlighted in bold and underlined. The in situ hybridization probe (isp) used for detection of fabp11b transcripts in adult zebrafish tissues, and PCR primers used for radiation hybrid map- ping (rhf, rhr) and for RT-PCR detection of fabp11b transcripts in RNA extracted from adult zebrafish tissues (rtf, rtr), are either underlined or overlined. AB CD Fig. 2. Spatiotemporal distribution of fabp11b transcripts during zebrafish embryonic and larval development was determined by whole mount in situ hybridization. fabp11b transcripts were first detected in the pigmented epithelium of the retina (Re) at 24 hpf (A). The distribution of fabp11b transcripts had spread across the retina by 30 hpf (B) and 48 hpf (C). In 5-day-old larvae, fabp11b transcripts were restricted to the circumference of the pigmented epithelium of the retina (D). S. Karanth et al. Duplicated fabp11 genes of zebrafish FEBS Journal 275 (2008) 3031–3040 ª 2008 The Authors Journal compilation ª 2008 FEBS 3033 eye of adult zebrafish (Fig. 4). fabp11b transcripts were not detected in total RNA extracted from the liver, intestine, kidneys, gills or muscle of adult zebrafish. As a positive control for the quality of RNA in each tissue, transcripts for the constitutively expressed ef1a gene were assayed by RT-PCR and detected in all tissues examined (Fig. 4). The difference observed in the tissue-specific distribution of fabp11b transcripts using in situ hybridization and RT-PCR is probably due to the sensitivities of the two assays; RT-PCR is far more sensitive than in situ hybridization. In con- trast to zebrafish fabp11b transcripts, fabp11a tran- scripts were detected in liver, intestine, brain, heart and muscle using RT-PCR [17]. Abundant fabp11 transcripts were detected in liver, adipose tissue and the vitellogenic ovary, transcripts were detected to a lesser extent in the previtellogenic ovary, heart, kidney, and muscle, and trace amounts were detected in testis by RT-PCR using total RNA extracted from tissues of adult Senegalese sole [12]. In adult Antarctic fishes [16], fabp11 (H ad -FABP) transcripts were detected in muscle, kidney, heart and brain by the less sensitive assay of northern blot and hybridization. Duplicate copies of fabp11 in zebrafish may have arisen by a fish-specific WGD event Multiple sequence alignments of selected zebrafish and mouse FABP amino acid sequences and the prototypic Fabp11 from the Senegalese sole [12] were performed using clustalw [23]. Zebrafish Fabp11b showed the highest sequence identity and similarity (65% and 84%, respectively) with zebrafish Fabp11a, and the next highest sequence identity and similarity with the Senegalese sole Fabp11 (63% and 82%, respectively) (Fig. 5). The sequence identity and similarity of zebra- fish Fabp11 decreased with paralogous FABP ⁄ Fabps from zebrafish and mouse (Fig. 5). Radiation hybrid mapping using the LN54 panel [24] assigned zebrafish fabp11b to LG (chromosome) 16 at a distance of 26.59 cR from the marker fc09b04 with a logarithm (base 10) of odds (LOD) score of 10. Zebrafish fabp11a had previously been assigned to LG (chromosome) 19 by the same LN54 radiation hybrid panel [17]. Conserved gene synteny on zebrafish LGs (chromosomes) 16 and 19 [17] with genes on human chromosome 8 (Table 1) suggest that fabp11a and fabp11b may have arisen by the teleost fish-specific WGD event that occurred approximately 250–400 mya [13]. Both fabp11a (LG 19) and fabp11b (LG 16), and two other duplicated genes on these LGs, showed conserved gene synteny with the FABP4–FABP5– FABP8–FABP9 gene cluster on human chromosome 8. For duplicated genes to be retained in the genome, Force et al. [25] have proposed that either both dupli- cated genes undergo subfunctionalization, in which the functions of the ancestral gene are subdivided between the sister duplicate genes, or one of the duplicates acquires a new function, called neofunctionalization. Force et al. [25] further proposed that subfunctional- ization of duplicated genes arises owing to the accumu- lation of mutations in the regulatory elements of duplicated genes, which leads to divergence in their tissue-specific patterns of expression. fabp11a mRNA transcripts were detected in ovary, liver, skin, intestine, brain, heart, testis and muscle in adult zebrafish [17]. During larval development, fabp11a transcripts were detected in the lens and diencephalon [17]. In contrast, fabp11b transcripts were detected in brain, heart, ovary AB Fig. 3. Tissue-specific distribution of fabp11b mRNA in adult zebra- fish sections determined by in situ hybridization. Sagittal (A) and transverse (B) sections of adult zebrafish were hybridized to an [ 33 P]dATP[aP] 3¢-end-labeled fabp11b antisense probe. The hybrid- ization signal of the antisense probe was limited to the spinal cord (Sc) of adult zebrafish. Fig. 4. RT-PCR detection of fabp11b transcripts in RNA extracted from adult tissues of zebrafish using fabp11b cDNA-specific prim- ers. fabp11b transcripts were detected by RT-PCR in RNA extracted from the brain (B), heart (H), ovary (O), and eye (E). No fabp11b transcripts were detected in RNA extracted from the liver (L), gills (G), intestine (I), muscle (M) or kidney (K) of adult zebrafish (upper panel). As a positive control, constitutively expressed elongation factor 1a (ef1a) transcripts were detected by RT-PCR in RNA extracted from all adult tissues examined (lower panel). Duplicated fabp11 genes of zebrafish S. Karanth et al. 3034 FEBS Journal 275 (2008) 3031–3040 ª 2008 The Authors Journal compilation ª 2008 FEBS and eye by RT-PCR (Fig. 4) and in the spinal chord by in situ hybridization (Fig. 3) in adult zebrafish, and in the retina during larval development (Fig. 2). Although fabp11a and fabp11b transcripts exhibit strik- ingly different patterns of tissue distribution in devel- oping embryos, larvae, and adult zebrafish, it is not possible to ascertain whether the duplicated copies of fabp11 have been retained in the zebrafish genome by either subfunctionalization or neofunctionalization, as there is no readily apparent ortholog of fish fabp11 in tetrapods or other species studied thus far [12]. fabp11, FABP4, FABP5, FABP8 and FABP9 evolved from a common ancestral gene on a single progenitor chromosome We constructed a neighbor-joining phylogenetic tree using the amino acid sequences of selected vertebrate FABPs (Fig. 6). Zebrafish Fabp11b and zebrafish Fabp11a formed a clade with other teleost Fabp11s (Fig. 6), a clade distinct from the frog, chicken and mammalian FABP4–FABP5–FABP8–FABP9 clade, as previously shown by Agulleiro et al. [12]. The zebrafish Fabp11a and Fabp11b amino acid sequences have diverged sufficiently from each other that they are not linked by a common node on the tree. Similarly, the putative fugu and stickleback Fabp11a and Fabp11b do not share a common node with their sister duplicates either, but are all clustered in the same clade with all other teleost fish Fabp11s. The sequence identity of zebrafish Fabp11b with mouse FABP4, FABP5, FABP8 and FABP9 (also known as PERF15) varied from 43% to 47% (Fig. 5). The sequence identity of the Senegalese sole Fabp11 with human FABP4, FABP5, FABP8 and FABP9 varied from 52.2% to 54.5% [12]. To date, Fig. 5. Zebrafish (D. rerio, Dr) Fabp11b is aligned with: zebrafish Fabp11a (deduced from AY628221), Fabp3 (AAL40832), Fabp7a (AAH55621) and Fabp7b (AAQ92970); Senegalese sole (Solea senegalensis, Sos) Fabp11 (CAM58515); mouse (M. musculus, Mm) FABP4 (AAH02148), FABP5 (NP_034764), FABP8 (XP_485204), and FABP9 (NP_035728). Dashes specify gaps and dots indicate amino acid identity. The percentage sequence identity and similarity of zebrafish, Senegalese sole and mouse FABP sequences with zebrafish Fabp11b are shown at the end of each sequence. S. Karanth et al. Duplicated fabp11 genes of zebrafish FEBS Journal 275 (2008) 3031–3040 ª 2008 The Authors Journal compilation ª 2008 FEBS 3035 we have no evidence for orthologs of tetrapod FABP4, FABP5, FABP8 and FABP9 in teleost fishes. This is based on tblastn searches of com- plete or nearly complete sequences for several fish genomes (e.g. Danio rerio, Takifugu rubrifus and Tet- raodon nigroviridis)inensembl (http://www.ensembl. org) and in the extensive expressed sequence tag (EST) databases (e.g. salmonids) in GenBank [26]. Agulleiro et al. [12] were the first to suggest that fabp11 formed a novel clade among vertebrate FABPs and that this gene is unique to teleost fishes. In addition to fabp11 of the Senegalese sole, homol- ogous fabp genes from the Antarctic fishes [16], and the zebrafish fabp genes reported here and previously [17], appear to belong to a sister group of FABP4, FABP5, FABP8 and FABP9 of mammals, and tetra- pod FABP4, FABP5, FABP8 and FABP9, and the teleost fish fabp11 genes might have arisen from a common ancestral gene [12]. On the basis of the evidence obtained from phylo- genetic analysis, linkage mapping and conserved gene synteny, we propose the following model for the evolution of fabp11 in fishes and the FABP4– FABP5–FABP8–FABP9 gene cluster in tetrapods (Fig. 7). First, fabp4, fabp5, fabp8 and fabp9 are absent in teleost fishes. Second, an ancestral gene diverged to give rise to the progenitors of fabp11 in fishes and FABP4, FABP5 , FABP8 and FABP9 in tetrapods, probably before the fish–tetrapod split some 450 mya [27]. Third, the FABP4–FABP5– FABP8–FABP9 gene cluster in tetrapods arose by successive tandem duplications and divergence of a single ancestral gene, as has been suggested for the evolution of some of the globin gene clusters [28]. Last, the ancestral gene of fabp11 and the FABP4– FABP5–FABP8–FABP9 gene cluster resided on a chromosome, or a part of it, that was the progenitor of zebrafish LGs (chromosomes) 16 and 19, chicken chromosome 2, mouse chromosome 3, human chro- mosome 8, and frog scaffold 225 (see Experimental procedures for details of retrieval of genomic sequence data). Table 1. Conserved gene synteny of the duplicated copies of zebrafish fabp11 with human FABP4 , FABP5, FABP8 and FABP9. Gene name Zebrafish Human Gene symbol LG position (cM) Mapping panel Gene symbol Chromosomal location Fatty acid-binding protein 11a fabp11a 19, 50.80–53.10 LN54 – – Fatty acid-binding protein 11b fabp11b 16, 44.16–45.60 LN54 – – Fatty acid-binding protein 4 – – – FABP4 8q21 Fatty acid-binding protein 5 – – – FABP5 8q21 Fatty acid-binding protein 8 – – – FABP8 8q21–8q22 Fatty acid-binding protein 9 – – – FABP9 8q21 der1-like domain family, member 1 derl1 16, 46.90 LN54 DERL1 8q24.13 NADH dehydrogenase (ubiquinone)-1beta subcomplex 9 ndufb9 16, 24.20 T51 NDUFB9 8q13.3 ATPase family, AAA containing 2 atad2 16, 26.70 T51 ATAD2 8q24.13 TAF2 RNA polymerase II, TATA box-binding protein (TBP)-associated factor taf2 16, 31.25 LN54 TAF2 8q24.12 RAD21 homolog rad21 16, 61.07 LN54 RAD2 8q24 Hyaluronan synthase 2 has2 16, 31.67 T51 HAS2 8q24.12 Cadherin 17 cdh17 16, 46.90 T51 CDH17 8q22.1 E3 ubiquitin protein ligase edd1 16, 67.71 T51 EDD1 8q22 Eukaryotic translation initiation factor 3, subunit 3 (gamma) eif3s3 16, 61.23 HS EIF3S3 8q24.11 Mitochondrial folate transporter ⁄ carrier mftc 19, 46.86 LN54 MFTC 8q22.3 Growth differentiation factor 6 gdf6b 19, 47.30 HS GDF6 8q22.1 Sperm-associated antigen 1 spag1 19, 49.00 T51 SPAG1 8q22.2 Protein tyrosine phosphatase type IVA, member 3 zfpm2b 19, 50.25 HS PTP4A3 8q24.3 Ribosomal protein L30 rpl30 19, 50.80 LN54 RPL30 8q22 Serine ⁄ threonine kinase 3 stk3 19, 51.88 T51 STK3 8q22.2 Metadherin lyric1 19, 51.95 T51 MTDH (LYRIC) 8q22.1 Antizyme inhibitor 1 azin1 19, 53.30 T51 AZIN1 8q22.2 Angiopoietin 1 angpt1 19, 81.93 T51 ANGPT1 8q22.3–q23 Duplicated fabp11 genes of zebrafish S. Karanth et al. 3036 FEBS Journal 275 (2008) 3031–3040 ª 2008 The Authors Journal compilation ª 2008 FEBS Tetrapod FABP5, FABP8, FABP9 and FABP4 are tandemly arrayed and probably arose by unequal crossing-over – which gene was duplicated first? Although it is speculative, it is possible to provide a parsimonious scheme for the tandem duplication events that gave rise to the cluster of four FABP genes on a single mammalian chromosome. In the frog genome, fabp4, fabp4-like and fabp5 are tan- demly arrayed on scaffold 225 (Fig. 7). Owing to sequence identity and their relationship in the phylo- genetic tree (Fig. 6), fabp4 and fabp4-like may have arisen in the frog genome as a result of tandem duplication of an ancestral gene on this chromo- some. It is not readily apparent, however, whether the tandem duplication that gave rise to frog fabp5 occurred before or after the tandem duplication that produced fabp4 and fabp4-like . The phylogenetic analysis would favor the fabp4⁄ fabp4-like duplication occurring after the duplication event that generated fabp5. Orthologs of mammalian FABP8 and FABP9 have not yet been identified, or more likely are not present, in the frog genome. Chicken FABP5, FABP8 and FABP4 are also tan- demly arrayed on chromosome 2; on the basis of data- base searches, it seems that FABP9 is absent from the chicken genome. Therefore, chicken FABP8 most probably arose from the tandem duplication of either FABP4 or FABP5. Again, the phylogenetic tree (Fig. 6) suggests that chicken FABP8 may have origi- nated from tandem duplication of FABP4 rather than FABP5,asFABP8 and FABP4 form a common clade, whereas FABP5 was placed in a different clade. Our model is consistent with the time-scale for these dupli- cation events based on synonymous ⁄ nonsynonymous amino acid substitution rates in FABP4, FABP5, FABP8 and FABP9, and the topology of the phylo- genetic tree reported by Schaap et al. [9]. Fig. 6. Phylogenetic tree of selected vertebrate FABPs, showing the relationship between zebrafish Fabp11a and Fabp11b. The neighbor-joining tree was constructed using H. sapiens LCN1 (NP_002288) as outgroup. The bootstrap values (per 100 duplicates) are indicated above or under each node. The teleost Fabp11s formed a common clade, which is indicated by a bracket. Amino acid sequences used in this analysis include: D. rerio (zebrafish) (Dr) Fabp11a (derived from AY628221), and Fabp11b (ENS- DARP0000002311); Gobionotothen gibberifrons (Gog) Fabp11 (H6- Fabp, AAC60354); Notothenia coriiceps (Nc) Fabp11 (H6-Fabp, AAC60352); Parachaenichthys charcoti (Pc) Fabp11 (H6- Fabp, AAC60355); Ta. rubripes (takifugu) (Fr) Fabp11a (deduced from AL837220), and Fabp11b (deduced from AL836636); Te. nigroviridis (Tn) Fabp11 (deduced from CR723700); Oryzias latipes (medaka) (Ol) Fabp11 (deduced from BJ899828); Cyprinus carpio (common carp) (Cc) Fabp11 (deduced from CF661735); So. senegalensis (Senegalese sole) (Sos) Fabp11 (CAM58515); Gasterosteus aculea- tus (stickleback) (Ga) putative Fabp11a (ENSGACP00000004532), and putative Fabp11b (ENSGACP00000011538); H. sapiens (Hs) FABP4 (CAG33184), FABP5 (AAH70303), and FABP8 (AAH34997); FABP9 (PERF15) (Uniprot ID Q0Z7S8); M. musculus (mouse) (Mm) FABP4 (AAH02148), FABP5 (NP_034764), FABP8 (XP_485204), and FABP9 (NP_035728); Rattus norvegicus (rat) (Rn) FABP4 (NP_445817), FABP5 (NP_665885), and FABP9 (NP_074045); Sus - scrofa (pig) (Ss) FABP4 (CAC95166); Gal. gallus (chicken) (Gg) FABP4 (NP_989621), FABP5 (ENSGALP00000025375), and FABP8 (ENSGALP00000025382); X. tropicalis (African frog) (Xt), putative FABP4 (ENSXETP00000022878), putative FABP4-like (NP_ 001096256.1), and putative FABP5 (ENSXETP00000022879); Pan troglodytes (chimpanzee) (Pt) FABP9 (PERF15) (ENS- PTRP00000053126). S. Karanth et al. Duplicated fabp11 genes of zebrafish FEBS Journal 275 (2008) 3031–3040 ª 2008 The Authors Journal compilation ª 2008 FEBS 3037 Fabp5, Fabp8, Fabp9 and Fabp4 are arranged in sequential order on mouse chromosome 3, and the FABP5–FABP8–FABP9–FABP4 gene cluster is present on human chromosome 8. Tandem duplication of FABP8 may have resulted in the formation of FABP9, as FABP8 and FABP9 formed a common clade (Fig. 6). Experimental procedures Husbandry of zebrafish Adult zebrafish were purchased from a local aquarium store and maintained according to established procedures [29]. Experimental protocols were reviewed by the Animal Care Committee of Dalhousie University in accordance with the Canadian Committee on Animal Care. Nucleotide sequence of zebrafish fabp11b cDNA and gene We retrieved a novel ensembl gene (ENSDARG000000023 11) from a blastn search of the zebrafish genome sequence database at the Wellcome Trust Sanger Institute (ver- sion Zv6; http://www.ensembl.org/Danio_rerio/index.html), using the fabp11a cDNA sequence as a query [17]. The novel ensembl gene (ENSDARG00000002311) exhibited most sequence identity and similarity to zebrafish fabp11a, hereafter referred to as zebrafish fabp11b. We confirmed the sequence of the coding region for fabp11b by comparison with fabp11b ESTs obtained from the blastn search of GenBank at the National Center for Biotechnology Infor- mation. All ESTs were identical to the coding region of zebrafish fabp11b. Phylogenetic analysis blosum62 matrix and clustalw [23] were used to align FABP sequences from zebrafish and other vertebrates. The bootstrap neighbor-joining tree was constructed using mega4 software [30]. Human LCN1 (NP_002288) was used as the outgroup. Radiation hybrid mapping of zebrafish fabp11b A detailed protocol for radiation hybrid mapping of zebra- fish genes is described by Hukriede et al. [24]. PCR reactions were carried out using the forward primer rhf (5¢-GT GTTGTGATTTTCGGTGG-3¢; nucleotide positions 33–51), and the reverse primer rhr (5¢-TTCTGTCATCTGCTG TCGTC-3¢; nucleotide positions 396–423), as shown in Fig. 1. PCR conditions were initial denaturation at 94 °C for 2 min, followed by 30 cycles at 94 °C for 30 s (denaturation), 54.5 °C for 30 s (primer annealing) and 72 °C for 1 min (elongation), with a final elongation step at 72 °C for 5 min. In situ hybridization to whole mount embryos and larvae, and to sections of adult zebrafish Whole mount in situ hybridization to zebrafish embryos and larvae was performed using a riboprobe synthesized Fig. 7. Evolutionary relationship between the duplicated copies of fabp11 in fishes and FABP4, FABP5, FABP8 and FABP9 (PERF15) in frog, chicken, mouse, and human. In zebrafish, fabp11a is found on LG 19, and fabp11b is on LG 16. In frog (X. tropicalis), fabp4, fabp4-like (fabp4l) and fabp5 are found on scaffold 225, where fabp4l and fabp5 are immediately adjacent to each other. Chicken (Gal. gallus) FABP5, FABP8 and FABP4 are located next to each other on chromosome 2. FABP5, FABP8, FABP9 (PERF15) and FABP4 are tandemly arrayed on human (H. sapiens) chromosome 8 and mouse (M. musculus) chromosome 3. Gaps in white indicate the presence of an additional gene between two FABP genes on a particular chromosome. Duplicated fabp11 genes of zebrafish S. Karanth et al. 3038 FEBS Journal 275 (2008) 3031–3040 ª 2008 The Authors Journal compilation ª 2008 FEBS from fabp11b cDNA clone zgc:110029 (GenBank accession number BC095142), as described by Thisse & Thisse [31]. In situ hybridization of an oligonucleotide probe to sections of adult zebrafish followed the protocol of Denovan-Wright et al. [32]. Briefly, sagittal and transverse sections of adult zebrafish were hybridized to [ 33 P]dATP[aP] 3¢-end-labeled fabp11b antisense probe, isf (5¢-CACAACACAAGACGTT TGACAGATAATAGC-3¢; nucleotide positions 11–40), shown in Fig. 1. Following hybridization and autoradio- graphy, tissue sections were stained with cresyl violet to identify specific tissues. RT-PCR detection of fabp11b transcripts in adult zebrafish tissues RT-PCR was employed for the tissue-specific detection of fabp11b transcripts in RNA extracted from tissues of adult zebrafish. Following synthesis of cDNA from RNA samples using the Omniscript RT kit (Qiagen, Mississauga, Ontario, Canada), fabp11b cDNA was PCR-amplified by the for- ward primer rtf (5¢-GCTGTCACTACATTCAA GACCCTGGA-3¢; nucleotide positions 337–363) and the reverse primer rtr (5¢-ACCATCCGCAAGGCTCATAGTA GT-3¢; nucleotide positions 1369–1393), shown in Fig. 1. PCR conditions for the amplification of fabp11b transcripts comprised an initial denaturation step at 94 °C for 2 min, followed by 30 cycles at 94 °C for 30 s (denaturation), 56 °C for 30 s (primer annealing) and 72 °C for 1 min (elongation), with a final elongation step at 72 °C for 5 min. PCR primers used for detection of elongation fac- tor 1a (ef1a) transcripts in total zebrafish RNA are described in Pattyn et al. [33]. The PCR conditions were an initial denaturation step at 94 °C for 2 min, followed by 30 cycles at 94 °C for 30 s (denaturation), 62 °C for 30 s (primer annealing) and 72 °C for 1 min (elongation), with a final elongation step of 72 °C for 5 min. Database searches of genome sequences and identification of transcription factor-binding sites The genomic organization of tetrapod FABP4, FABP5, FABP8 and FABP9 was derived from the Xenopus tropical- is, Gallus gallus, Mus musculus and Homo sapiens genome sequence databases at http://www.ensembl.org. Putative transcription factor-binding sites in the 5¢-upstream region of zebrafish fabp11b were identified using alibaba 2.1 soft- ware [34]. Acknowledgements The authors thank Mark Soric and Fernanda Alves- Costa for technical assistance, and David R. Smith and Dr Tudor Borza for helpful comments. This work was supported by funds from the Natural Sciences and Engineering Research Council of Canada (to J. M. Wright), Canadian Institutes of Health Research (to E. Denovan-Wright), and the National Institutes of Health, the European Commission as part of the ZF-Models integrated project in the 6th Framework Programme (to B. Thisse and C. Thisse). S. Karanth is a recipient of a Faculty of Graduate Studies Scholar- ship from Dalhousie University. References 1 Bernlohr DA, Simpson MA, Hertzel AV & Banaszak LJ (1997) Intracellular lipid-binding proteins and their genes. Annu Rev Nutr 17, 277–303. 2 Corsico B, Liou HL & Storch J (2004) The alpha- helical domain of liver fatty acid binding protein is responsible for the diffusion-mediated transfer of fatty acids to phospholipid membranes. Biochemistry 43, 3600–3607. 3 Ho SY, Delgado L & Storch J (2002) Monoacylglycerol metabolism in human intestinal Caco-2 cells: evidence for metabolic compartmentation and hydrolysis. J Biol Chem 277, 1816–1823. 4 Murota K & Storch J (2005) Uptake of micellar long- chain fatty acid and sn-2-monoacylglycerol into human intestinal Caco-2 cells exhibits characteristics of protein- mediated transport. J Nutr 135, 1626–1630. 5 Storch J, Veerkamp JH & Hsu KT (2002) Similar mech- anisms of fatty acid transfer from human and rodent fatty acid-binding proteins to membranes: liver, intes- tine, heart muscle and adipose tissue FABPs. Mol Cell Biochem 239, 25–33. 6 Veerkamp JH & van Moerkerk HTB (1993) Fatty acid- binding protein and its relation to fatty acid oxidation. Mol Cell Biochem 123, 101–106. 7 Ong DE, Newcomer ME & Chytil F (1994) Cellular retinoid-binding proteins. In The Retinoids: Biology, Chemistry and Medicine, 2nd edn. (Sporn MB, Roberts AB & Goodman DS, eds), pp. 283–317. Raven Press, New York, NY. 8 Hertzel AV & Bernlohr DA (2000) The mammalian fatty acid-binding protein multigene family: molecular and genetic insights into function. Trends Endocrine Metab 11, 175–180. 9 Schaap FG, van der Vusse GJ & Glatz JFC (2002) Evo- lution of the family of intracellular lipid binding pro- teins in vertebrates. Mol Cell Biochem 239, 69–77. 10 Schleicher CH, Co ´ rdoba OL, Santome ´ JA & Dell’Angelica EC (1995) Molecular evolution of the multigene family of intracellular lipid-binding proteins. Biochem Mol Biol Int 36, 1117–1125. 11 Chan L, Wei C-F, Li W-H, Yang C-Y, Ratner P, Pownall H, Gotto AM Jr & Smith LC (1985) Human liver fatty acid binding protein cDNA and amino acid S. Karanth et al. Duplicated fabp11 genes of zebrafish FEBS Journal 275 (2008) 3031–3040 ª 2008 The Authors Journal compilation ª 2008 FEBS 3039 sequence. Functional and evolutionary implications. J Biol Chem 260 , 2629–2632. 12 Agulleiro MJ, Andre ´ M, Morais S, Cerda ` J & Babin PJ (2007) High transcript level of fatty acid-binding pro- tein 11 but not of very low-density lipoprotein receptor is correlated to ovarian follicle atresia in a teleost fish (Solea senegalensis). Biol Reprod 77, 504–516. 13 Furlong RF & Holland PW (2002) Were vertebrates octoploid? Phil Trans R Soc Lond B Biol Sci 357, 531– 544. 14 Ohno S (1970) Evolution by Gene Duplication . Springer, New York, NY. 15 Vandepoele K, De Vos W, Taylor JS, Meyer A & Van de Peer Y (2004) Major events in the genome evolution of vertebrates: paranome age and size differ considerably between ray-finned fishes and land vertebrates. Proc Natl Acad Sci USA 101, 1638– 1643. 16 Vayda ME, Londraville RL, Cashon RE, Costello L & Sidell BD (1998) Two distinct types of fatty acid-bind- ing protein are expressed in heart ventricle of Antarctic teleost fishes. Biochem J 330, 375–382. 17 Liu R-Z, Saxena V, Sharma MK, Thisse C, Thisse B, Denovan-Wright EM & Wright JM (2007) The fabp4 gene of zebrafish ( Danio rerio) – genomic homology with the mammalian FABP4 and divergence from the zebrafish fabp3 in developmental expression. FEBS J 274, 1621–1633. 18 Breathnach R & Chambon P (1981) Organization and expression of eukaryotic split genes coding for proteins. Annu Rev Biochem 50, 349–383. 19 Divine JK, Staloch LJ, Haveri H, Rowley CW, Hei- kinheimo M & Simon TC (2006) Cooperative interac- tions among intestinal GATA factors in activating the rat liver fatty acid binding protein gene. Am J Physiol- Gastr L. 291, G297–306. 20 Sellner P (1994) Developmental regulation of fatty acid binding protein in neural tissue. Dev Dynam 200, 333– 339. 21 Liu R-Z, Denovan-Wright EM, Degrave A, Thisse C, Thisse B & Wright JM (2004) Differential expression of duplicated genes for brain-type fatty acid-binding pro- teins (fabp7a and fabp7b) during early development of the CNS in zebrafish (Danio rerio). Gene Expr Patterns 4, 379–387. 22 Narayanan V, Kaestner KH & Tennekoon GI (1991) Structure of the mouse myelin P2 protein gene. J Neu- rochem 57, 75–80. 23 Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F & Higgins DG (1997) The CLUSTAL_X windows interface: flexible strategies for multiple sequence align- ment aided by quality analysis tools. Nucleic Acids Res 25, 4876–4882. 24 Hukriede NA, Joly L, Tsang M, Miles J, Tellis P, Epstein JA, Barbazuk WB, Li FN, Paw B, Postlethwait JH et al. (1999) Radiation hybrid mapping of the zebra- fish genome. Proc Natl Acad Sci USA 96, 9745–9750. 25 Force A, Lynch M, Pickett FB, Amores A, Yan YL & Postlethwait JH (1999) Preservation of duplicate genes by complementary, degenerative mutations. Genetics 151, 1531–1545. 26 Benson DA, Karsch-Mizrachi I, Lipman DJ, Ostell J & Wheeler DL (2005) GenBank. Nucleic Acids Res 33, D34–D38. 27 Kumar S & Hedges SB (1998) A molecular timescale for vertebrate evolution. Nature 392, 917–920. 28 Hardison R (1998) Hemoglobins from bacteria to man: evolution of different patterns of gene expression. J Exp Biol 201, 1099–1117. 29 Westerfield M (2000) The Zebrafish Book: A Guide for the Laboratory Use of Zebrafish (Danio rerio). Univer- sity of Oregon Press, Eugene, OR. 30 Tamura K, Dudley J, Nei M & Kumar S (2007) MEGA4: molecular evolutionary genetics analysis (MEGA) soft- ware version 4.0. Mol Biol Evol 24, 1596–1599. 31 Thisse C & Thisse B (2008) High-resolution in situ hybridization to whole-mount zebrafish embryos. Nat Protoc 3, 59–69. 32 Denovan-Wright EM, Newton RA, Armstrong JM, Babity JM & Robertsin HA (1998) Acute administra- tion of cocaine, but not amphetamine, increase the level of synaptotgmin IV mRNA in the dorsal striatum of rat. Mol Brain Res 55, 350–354. 33 Pattyn F, Robbrecht P, Speleman F, De Paepe A & Vandesompele J (2006) RTPrimerDB: the real-time PCR primer and probe database, major update 2006. Nucleic Acids Res 34, D684–D688. 34 Grabe N (2002) Alibaba2: context specific identification of transcription factor binding sites. In Silico Biol 2, S1–S15. Duplicated fabp11 genes of zebrafish S. Karanth et al. 3040 FEBS Journal 275 (2008) 3031–3040 ª 2008 The Authors Journal compilation ª 2008 FEBS . following their divergence, gave rise to the fabp11 genes of zebrafish, and the progenitor of the FABP4, FABP5, FABP8 and FABP9 genes in tetrapods after the separation of the fish and tetrapod lineages. Abbreviations CRABP,. The evolutionary relationship between the duplicated copies of the zebrafish fabp11 gene and the tetrapod FABP4, FABP5, FABP8 and FABP9 genes Santhosh Karanth 1 , Eileen. of FABP4, FABP5, FABP8 and FABP9 of mammals, and tetra- pod FABP4, FABP5, FABP8 and FABP9, and the teleost fish fabp11 genes might have arisen from a common ancestral gene [12]. On the basis of