当前位置: 首页 > 医学版 > 期刊论文 > 基础医学 > 分子生物学进展 > 2004年 > 第1期 > 正文
编号:11259300
Identification of Duplicated Fourth 2-Adrenergic Receptor Subtype by Cloning and Mapping of Five Receptor Genes in Zebrafish
     * Department of Pharmacology and Clinical Pharmacology

    Turku Graduate School of Biomedical Sciences, University of Turku, Turku, Finland

    Department of Biochemistry and Pharmacy, ?bo Akademi University, Turku, Finland

    Department of Neuroscience, Unit of Pharmacology, Uppsala University, Uppsala, Sweden

    | Institute of Neuroscience, University of Oregon, Eugene, Oregon

    E-mail: mschein@utu.fi.

    Abstract

    The 2-adrenergic receptors (2-ARs) belong to the large family of rhodopsinlike G-protein–coupled receptors that share a common structure of seven transmembrane (TM) -helices. The aims of this study were (1) to determine the number of 2-AR genes in a teleost fish, the zebrafish (Danio rerio), (2) to study the gene duplication events that generated the 2-AR subtypes, and (3) to study changes in receptor structure that have occurred since the divergence of the mammalian and fish lineages. Here, we report the cloning and chromosomal mapping of fish orthologs for all three mammalian 2-ARs. In addition, we identified a fourth 2-AR subtype with two duplicates in zebrafish. Chromosomal mapping showed that the zebrafish 2-AR genes are located within conserved chromosomal segments, consistent with the origin of the four 2-AR subtypes by two rounds of chromosome or block duplication before the divergence of the ray fin fish and tetrapod lineages. Thus, the fourth subtype has apparently been present in the common ancestor of vertebrates but has been deleted or is yet to be identified in mammals. The overall percentage identity between the fish and mammalian orthologs is 53% to 67%, and in the TM regions 80% to 87%. These values are clearly lower than what is observed between mammalian orthologs. Still, all of the residues thought to be important for 2-adrenergic ligand binding are conserved across species and subtypes, and even the most divergent regions of the fish receptors show clear "molecular fingerprints" typical for orthologs of a given subtype.

    Key Words: genome duplication ? synteny ? fish ? 2-adrenergic receptor

    Introduction

    Adrenergic receptors (AR) mediate the physiological effects of the biogenic amine hormones/neurotransmitters adrenaline and noradrenaline. ARs can be divided into three main classes, 1-AR, 2-AR, and ?-AR, each of which is represented by three subtypes in humans and other mammals (1A, 1B, and 1D; 2A, 2B, and 2C; ?1, ?2, and ?3). ARs belong to a large family of cell surface receptors that control intracellular second messenger systems by activating guanine nucleotide–binding regulatory proteins (G-proteins). The G-protein coupled receptor (GPCR) superfamily is one of the largest gene superfamilies in animal genomes and comprises about 5% of all genes of the C. elegans genome and about 2% of the human genome (Bargmann 1998; Bockaert and Pin 1999; Venter et al. 2001). It is estimated that 40% to 50% of current therapeutic drugs act on GPCRs (Katugampola and Davenport 2003). The tertiary structure of these receptors is characterized by a prominent bundle of seven -helical regions of 23 to 33 predominantly hydrophobic amino acids that span the cell membrane (Palczewski et al. 2000).

    Three different intronless genes encode the three distinct mammalian 2-AR subtypes, 2A, 2B, and 2C. The human 2-AR genes have been designated ADRA2A (2C10), ADRA2B (2C2), and ADRA2C (2C4) and are located on human chromosomes 10, 2, and 4 (Kobilka et al. 1987; Regan et al. 1988; Lomasney et al. 1990; Bylund et al. 1992). Related sets of 2-AR subtype genes or cDNAs have been cloned from some other mammalian species used as experimental animals such as rat, mouse, and guinea pig (for review, see MacDonald, Kobilka, and Scheinin [1997]). The rodent 2A-AR was initially designated 2D-AR on the basis of its ligand-binding profile but was subsequently shown to be orthologous to human 2A-AR (Lanier et al. 1991; Link et al. 1992). Occasionally, it is still misleadingly called 2D. Despite this, we propose that the fourth 2-AR subtype reported in this paper should be called 2D-AR. There are two duplicates of this receptor subtype in zebrafish, encoded by the genes adra2da and adra2db.

    Single counterparts of the human 2-AR subtype genes have also been cloned from several mammalian species such as pig (2A) (Guyer et al. 1990), cow (2A) (Venkataraman, Duda, and Sharma 1997), tree shrew (2A and 2C, both are partial clones) (Meyer et al. 2000; Flugge et al. 2003), and opossum (2C) (Blaxall et al. 1994). Partial 2B-AR genes have also been cloned from a range of mammals for use in molecular phylogeny analysis and to study the molecular evolution of mammalian 2B-ARs (Springer et al. 1997; Stanhope et al. 1998; Madsen et al. 2001; Springer et al. 2001; Teeling et al. 2002; Madsen et al. 2002). A previous study provides evidence for the existence of different classes of adrenergic receptors in nonmammalian vertebrates, too. This cross-hybridization study, using different adrenergic receptor probes, identified sequences homologous to human 2A-AR in goldfish, a frog, a turtle, and chicken (Palacios et al. 1989). A partial cDNA sequence is known for a chicken 2A-AR (Blaxall, Heck, and Bylund 1993). The ligand-binding properties of the 2-AR cloned from the fish cuckoo wrasse (Labrus ossifagus) are intermediate between 2A and 2C, and it was initially thought to represent an "ancestral" 2-AR subtype (Svensson et al. 1993). On the amino acid sequence level, this "2F-AR" shows the greatest similarity to the 2C-AR subtype. In GenBank, there is a genomic DNA sequence of a putative full-length goldfish 2-AR gene (L09064, unpublished) that does not appear to be orthologous to any of the three mammalian 2-AR subtypes (MacDonald, Kobilka, and Scheinin 1997). In addition, in GenBank, there are fragments of nucleotide sequences of putative 2-AR subtypes from lamprey, several fish species, birds, a reptile, and an amphibian (accession numbers AL606538 to AL606586, unpublished). Eight different 2-AR protein sequences from the pufferfish (Takifugu rubribes), the first fish with a sequenced genome, have been recently listed in the SWISS-PROT protein database (Q8JG00 to Q8JG07, unpublished). Currently no 2-ARs have been definitively identified in invertebrates. Several biogenic amine receptors are predicted in the mosquito and fruit fly genomes (Hill et al. 2002). The fruit fly octopamine/tyramine and snail octopamine receptor are highly similar to the mammalian adrenergic receptors, and they also bind 2-adrenergic drugs (Gerhardt et al. 1997; Chatwin et al. 2003). A tyramine receptor in the nematode C. elegans has also been characterized (Rex and Komuniecki 2002). Several putative "alpha adrenergic" receptors have been annotated in the mosquito, fruit fly, and nematode genomes (genomes browseable via www.ensembl.org), although it is unclear whether any of these species have adrenaline or noradrenaline as natural ligands.

    The diversity of GPCR subtypes seems to have arisen in early vertebrate evolution and may thereby be explained by the extensive increase in gene number (Miyata and Suga 2001; Spring 2003) proposed to be the result of two genome duplications before the radiation of jawed vertebrates some 500 MYA (Holland et al. 1994; Vernier et al. 1995; Postlethwait et al. 1998). This evolutionary view is supported by the chromosomal localization of the human AR genes in related chromosomal segments called paralogy groups or paralogons (Pebusque et al. 1998; Wraith et al. 2000). The much-debated HOX-paralogons do not overlap with the ADR-bearing paralogons. It is noteworthy that the human chromosome Hsa2 is a result of a recent fusion of two different chromosomes in the human lineage, the ADRA2B-bearing Hsa2p having a different origin from the HOX-bearing Hsa2q (Larhammar, Lundin, and Hallbook 2002). From this point of view, our starting hypothesis was that the fish genome should have the same three 2-AR subtype genes as mammals (and possibly duplicates of some of these if the teleost fish genome has undergone an additional genome doubling followed by some gene loss [Postlethwait et al. 1998; Taylor et al. 2001]). In addition, a fourth subtype gene might be present, representing the expected fourth product of two sets of genome duplication, currently thought to be missing from mammalian genomes. To study the evolution and structural conservation of 2-ARs and to facilitate further studies on their possible importance in developmental biology, we turned to a genetically well-characterized species of fish, the zebrafish (Danio rerio), which is highly amenable to developmental and genetic studies and useful in studies on brain aminergic systems, too (Guo et al. 1999). Based on the conservation of the HOX-gene clusters across species, it is possible that there has been an extra round of chromosomal duplications in the teleost lineage, leading to two orthologs for a single mammalian gene (Amores et al. 1998; Postlethwait et al. 1998; Naruse et al. 2000). Recently, there has been a debate on the nature of the origin of the duplicated fish genes (i.e., whether they arose from local or large-scale duplications) (Taylor et al. 2001; Taylor, Van de Peer, and Meyer 2001; Robinson-Rechavi et al. 2001a; Robinson-Rechavi 2001b. The 2-ARs belong to another well-conserved gene cluster, and we wished to test the hypothesis that the evolutionary pattern displayed by the HOX-containing chromosome regions can be generalized to other regions of the zebrafish genome.

    Materials and Methods

    Library Screening and Isolation of Receptor Clones

    All probes were 32P-labeled using the MegaPrime labeling kit (Amersham Pharmacia Biotech UK, Ltd., Buckinghamshire, England). The coding sequence of the cloned cuckoo wrasse 2-AR (Svensson et al. 1993) was used as a probe to screen a Zap II -vector cDNA-library (prepared and provided by D. J. Grunwald, University of Utah) and a FIX II -vector genomic DNA-library (prepared and provided by Scott Stachel, University of California). Hybridizations were carried out in 25% formamide, 6 x SSC, 10% dextran sulphate, 5 x Denhardt's solution, and 0.1% SDS at 42°C overnight. The filters were washed twice in 2 x SSC/0.1% SDS at room temperature for 5 min and twice in 2 x SSC/0.1% SDS for 30 min at 42°C. One strongly positive clone from the cDNA-library was rescued using the helper phage R 408, obtained as a pBluescript II SK– clone, and sequenced. A truncated cDNA clone showing similarity to the cuckoo wrasse 2F-AR was obtained. Eighteen candidate clones from the genomic library were digested with a series of restriction enzymes, Southern blotted, and hybridized with the same probe. One positive clone was selected, and a band of suitable size was ligated into a pGEM-vector and sequenced. A genomic clone containing a 1,119-bp sequence coding for an 2A-like AR was obtained. This clone was truncated at its 3' end. One positive phage clone with a different restriction pattern was sequenced directly and found to be overlapping with this 2A-like clone with 100% identity. A complete ORF of 1,167 bp was deduced from the sequences of these two clones. Other clone candidates were tested using the polymerase chain reaction (PCR) and hybridizations and were found to represent either identical copies of this gene or false positives.

    Another zebrafish genomic DNA-library in a Lawrist7 cosmid-vector (from the Resource Center/Primary Database (RZPD), Max-Planck-Institute for Molecular Genetics, Berlin, Germany) was screened with the cDNA clone. Hybridization was carried out as above but at 65°C. The filters were washed twice in 2 x SSC/0.1% SDS for 5 min at 65°C and twice in 0.1 x SSC/0.1% SDS for 30 min at 65°C. A clone containing a 1,299-bp ORF coding for an 2C-like AR was isolated as above, as well as two clones containing identical sequences. The same library was screened for additional subtypes using a probe corresponding to a zebrafish 2B-AR like EST (AI461341, coding for transmembrane [TM] domains 6 and 7 and the carboxyl-terminal tail). This hybridization was carried out in ExpressHyb-buffer (Clontech, Palo Alto, Calif.) for 2 h at 60°C. The filters were washed twice in 2 x SSC/0.1% SDS for 5 min at 60°C, twice in 0.5 x SSC/0.1% SDS for 30 min, and once in 0.5 x SSC/0.1% SDS for 15 min at 60°C. Sixteen positive cosmid clones were isolated and analyzed as above. One clone contained a 1,533-bp ORF coding for an 2B-like AR. Another cosmid clone was directly sequenced further using primer walking. This clone contained a truncated coding sequence for an 2-AR differing from the other zebrafish 2-ARs. The cosmid clone was truncated at its 5' end, lacking the N-terminus and N-terminal half of TM1. A Blast search with this truncated sequence identified a closely related, but not identical, truncated 2-AR sequence (GenBank accession number AL606583). AL606583 spans from the C-terminal end of TM3 to the second half of TM6 (ending with conserved sequence CWF). The other clones contained copies of related genes (see Results) or were false positives.

    Inverse PCR

    Zebrafish genomic DNA was prepared as described previously (Akimenko 1995). One μg DNA/reaction was digested with eight different restriction enzymes, BamHI, EcoRI, HindIII, KpnI, PstI, SacI, SpeI, and XhoI (New England Biolabs Inc., Beverly, Mass.), overnight at 37°C. The digests were diluted to 350 μl and extracted with an equal volume of phenol:chloroform (1:1). After gentle vortexing (20 s) and centrifugation, the aqueous phase containing the DNA was removed. DNA was precipitated with 0.05 volumes of 3 M sodium acetate and 2 volumes of cold 99% ethanol and centrifugation at 12,000 rpm for 15 min. The precipitate was dissolved in 449 μl nuclease-free water. Ligase buffer, 50 μl, and 1 μl (400 U) T4-ligase (New England Biolabs) were added, and ligation was carried out overnight at 14°C. DNA was precipitated with 150 μl 5 M ammonium acetate and 700 μl 99% ethanol, centrifuged, and washed with 70% ethanol. The pellet was dissolved in 10 μl of nuclease-free water.

    Two sets of primers were designed based on the obtained truncated coding sequences from the cosmid clone and the related genomic sequence (AL606583), one inner primer pair and one outer pair for each of these sequences. The primers were synthesized in the reverse complementary direction to amplify the nucleotide sequences upstream and downstream from the known sequences. The primer sequences are provided in the Supplementary Material online.

    PCR reactions were run using a DynaZyme EXT kit with proofreading DNA polymerase (FinnZymes, Espoo, Finland) and the following reagents: forward primer 15 pmol, reverse primer 15 pmol, and DMSO 1:20. Primary PCR reactions were run with the inner primer pair and 1 μl of the ligation products as templates, using the following cycling conditions: 94°C for 2 min, 1 cycle; 94°C for 30 s, 53°C for 30 s, and 72°C for 2 min, 40 cycles, and final extension at 72°C for 7 min. One μl of the primary PCR reactions was used as the template in a second PCR using the outer primer pair (conditions as above, except with an annealing temperature of 54°C; for AL606583 the annealing temperature was 59°C for both reactions). The second PCR reactions were analyzed on a 1.5% agarose gel and single bands were identified from the gel, excised, and purified using the High Pure PCR Product Purification Kit (Roche Molecular Biochemicals, Mannheim, Germany). The fragments were either directly sequenced using the outer PCR primers or A/T-ligated to pGEM-T-Easy vector (Promega, Madison, Wis.) according to the manufacturer's instructions, introduced into XL-1 Blue supercompetent cells (Stratagene, La Jolla, Calif.) for plasmid growth, and isolated using the NucleoSpin Plasmid kit (MACHEREY-NAGEL GmbH & Co., Düren, Germany). The plasmid preparations were checked for correct inserts using restriction digestions and subsequently sequenced with the universal primers T7 and SP6.

    For the cosmid clone, one of the fragments (EcoRI-digested initial template) contained a 190-bp sequence overlapping with the cosmid clone sequence and 195 bp of good quality upstream sequence, the overlapping parts of the sequences being 100% identical. The nucleotide sequence containing a 1,227-bp ORF for a previously unpublished zebrafish 2-AR (2Da-AR) was deduced from the sequences of the truncated cosmid clone and the inverse-PCR product.

    For AL606583, the full-length coding sequence was constructed in a similar way. Fragments amplified from SacI-digested and EcoRI-digested initial templates contained an upstream sequence with a start codon and 287 bp of 5'-UTR. A fragment from the BamHI-digested initial template contained the downstream sequence, a stop codon, and 38 bp of 3'-UTR. The overlapping parts of the sequences were 100% identical. By combining the sequences of AL606583 and the inverse PCR–generated fragments, another previously unpublished zebrafish 2-AR gene sequence containing a 1,248-bp ORF was obtained (2Db-AR). The obtained gene sequences were also used to search the Sanger Institute's zebrafish genome database (www.sanger.co.uk, available via Blast and keywords from www.ensembl.org/Danio_rerio/) to identify possible additional 2-AR sequences and to check for possible sequence variants of the identified genes.

    DNA Sequencing and Sequence Analysis

    All sequencing was performed using the ABI Prism Automatic sequencer (Applied Biosystems Inc., Foster City, Calif.) and several specific oligonucleotide primers on both DNA strands. Blast searches (from http://www.ncbi.nlm.nih.gov:80/BLAST/ for GenBank entries and from www.ensembl.org for zebrafish genome specific sequences, GENSCAN predictions) using full-length DNA-DNA (BlastN) or protein (BlastP) (sequences spanning TM1 to TM5 and TM6 to TM7) against translated databases; keyword searches for annotated sequences were used to identify related sequences. Raw sequence assembly and more detailed sequence analysis were performed using the Lasergene software package (DNASTAR Inc., Madison, Wis.). Sequences from other species were retrieved from GenBank and nucleotide sequences were translated into putative protein sequences using the Lasergene package. As a result, in addition to the five cloned zebrafish 2-AR receptors, 27 full-length and 46 truncated protein sequences of putative 2-ARs were found. Multiple sequence alignments were constructed using Malign (Johnson and Overington 1993) for (1) the full-length sequences, (2) the regions common to all of the 78 available 2-AR sequences, and (3) separately for sequences from each of the four 2-AR subtypes. These individual alignments were themselves aligned, modified manually over the highly divergent regions, and are presented in the Supplementary Material online. Only unambiguously aligned regions from these alignments were used for molecular phylogeny analysis. Phylogenetic trees were constructed with the Neighbor-Joining and maximum-parsimony methods using the programs SEQBOOT, PROTPARS, PROTDIST, Neighbor, and Consense from the PHYLIP version 3.5c package (Felsenstein 1993) and with the maximum-likelihood method using PROTML form the MOLPHY package. The statistical significance of the presented branching was determined using bootstrap analysis (1,000 data replicates). In contrast to trees based on the ortholog-specific alignments, the tree based on the sequences common to all of the 78 available 2-AR sequences (data not shown) showed low stability, as reflected in the bootstrap analysis, probably because there was an insufficient number of phylogenetically informative sites.

    RT-PCR

    Total RNA was isolated either from a pool of 30 zebrafish embryos (age 48 to 50 hpf) or an individual whole adult zebrafish using RNAwiz (Ambion, Inc., Austin, Tex.). The RNA samples were DNase treated using DNA-free (Ambion) to remove contaminating genomic DNA. Specific primers for adra2da and adra2db were first tested for performance using genomic DNA as a template. The primers were also checked for specificity using plasmids containing the coding sequence for the other duplicate as a template; no cross-recognition was detected. Primer sequences are provided in the Supplementary Material online. In RT-PCR, 100 ng of total RNA was used as the template for the Access RT-PCR System (Promega) with only slight modifications to the manufacturer's recommendations. The final Mg2+-concentration used was 1.5 mM. First-strand cDNA synthesis was performed at 48°C for 45 min followed by RT-enzyme inactivation at 94°C for 2 min. Second-strand synthesis and PCR amplifications were performed using the following cycling conditions: 94°C for 30 s, 60°C for 1 min, and 68°C for 1 min, 40 cycles, and final extension at 68°C for 7 min. Control reactions were run without the RT-enzyme to detect the possible existence of contaminating genomic DNA.

    Southern Hybridizations

    Genomic DNA from individual, whole adult zebrafish was prepared as described previously (Akimenko 1995), digested with HindIII restriction enzyme, run on a gel, and blotted on a nylon filter (Amersham Pharmacia Biotech UK). The cloned zebrafish 2-AR genes do not contain HindIII recognition sites within their coding sequences. The same 2B-AR and 2C-AR probes, as well as a zebrafish 2A-AR probe corresponding to TM6, TM7, and surrounding regions, were used as for the initial screenings. The 2Da-AR and 2Db-AR probes correspond to the N-terminal part of the receptor (TM1 to TM4 and surrounding regions). Southern hybridizations of filters were performed while gradually lowering the stringency. All hybridizations were carried out in 25% formamide, 6 x SSC, 10% dextran sulphate, 5 x Denhardt's solution, and 0.1% SDS at 65°C, 55°C, 50°C, or 42°C overnight. The filters were first washed twice in 2 x SSC/0.1% SDS at room temperature for 5 min and, depending on the stringency, twice in 0.2 x SSC/0.1% SDS for 30 min at 65°C (high stringency), twice in 0.5 x SSC/0.1% SDS for 30 min at 55°C (high to intermediate), twice in 0.5 x SSC/0.1% SDS for 30 min at 50°C (intermediate to low) or twice in 2 x SSC/0.1% SDS for 30 min at 42°C (low). After each washing step, the filters were exposed either with films or imaging plates, analyzed with a Fuji PhosphorImager (Fuji Corporation, Tokyo, Japan), and rehybridized.

    Chromosomal Mapping of the Cloned Genes and Genome Comparisons

    PCR primers were designed to amplify a region near the coding sequences of the cloned zebrafish 2-AR genes (adra2a, adra2b, adra2c, adra2da, and adra2db) containing single-strand conformation polymorphisms. Primer sequences for the different genes are provided in the Supplementary Material online. Primers were first tested for performance with genomic zebrafish DNA as the template. The detected polymorphisms were scored on the HS (heat shock) meiotic mapping panel and linkage analysis was performed to determine their chromosomal location. The method is described in detail elsewhere (Kelly et al. 2000; Woods et al. 2000). Gene locations for chromosomal comparisons between different species were retrieved from NCBI's HomoloGene service (http://www.ncbi.nlm.nih.gov/HomoloGene/) and The Zebrafish Server (http://zfin.org/).

    Results

    Library Screening and Isolation of Receptor Clones

    One cDNA and two genomic libraries were screened under different stringencies to obtain the zebrafish 2-AR genes. A truncated cDNA clone showing a high degree of sequence identity with the cuckoo wrasse 2-AR was isolated from the cDNA library. A genomic DNA library was screened with this cDNA clone under high stringency hybridization conditions. A clone containing a 1,296-bp ORF coding for a zebrafish 2C-AR (gene name adra2c) was isolated, as well as two clones containing identical sequences. The same library was screened for additional 2-AR subtypes using an EST-derived, PCR-generated 2B-AR–resembling probe under low stringency hybridization conditions. Sixteen clones were isolated and analyzed. One clone contained a 1,533-bp ORF coding for a zebrafish 2B-AR (gene name adra2b). Four clones were found to contain sequences identical to the zebrafish 2C-AR, and two clones contained identical or almost identical sequences of a previously identified zebrafish μ-opioid receptor (AF132813) (Barrallo et al. 2000), as indicated by a simple Blast search (data not shown). One clone contained a truncated sequence for a previously unpublished subtype, 2Da-AR (gene name adra2da). The full-length 1,227-bp sequence for 2Da-AR was obtained using inverse-PCR; that is, amplifying sequences upstream and downstream from the known sequence using PCR primers constructed in reverse complement order with circularized genomic DNA as template (see the Materials and Methods). The remaining clones turned out to be false positives.

    A nucleotide fragment (GenBank accession number AL606583) with high similarity to the 2Da-AR sequence was found using a Blast search. The full-length 1,248-bp sequence for this receptor, the 2Db-AR (gene name adra2db), was subsequently obtained using inverse PCR.

    A genomic clone containing a 1,119-bp coding sequence for a zebrafish 2A-AR (gene name adra2a) was isolated from the other genomic DNA library and hybridized with the cuckoo wrasse 2-AR probe under low stringency conditions. The coding sequence was interrupted 9 amino acids after TM7. Another phage clone was sequenced to obtain the complete 1,167-bp ORF of this gene and showed that the carboxyl-terminal tail consists of 24 amino acids, similarly to the mammalian 2A-ARs (all of equal length).

    Blast searches with the cloned zebrafish 2-AR sequences resulted in two exact or almost exact matches with ESTs: AW077216 overlaps the N-terminal part of adra2a and has one nucleotide difference leading to one amino acid change in TM1 (I27V, not facing the binding cavity according to the modeled structure). The other EST, AI461341, is 100% identical to the zebrafish 2B-AR over the overlapping sequence at the C-terminus. Other exact or nearly exact matches from GenBank include genomic fragments of 2B-AR (AL606585, 99.1% identical, including one amino acid deletion in the third intracellular loop), 2C-AR (AL606584, 99.2% identical, no amino acid differences), 2Da-AR (AL606586, 98.7% identical, one amino acid difference in the third intracellular loop), and 2Db-AR (AL606583, 99.9% identical, no amino acid differences). The quality of these sequences is not easy to assess, as the method of isolation and sequencing of these entries has not been reported. For this reason, we relied on our own adra2 sequences in the phylogenetic analyses. A 99.5% identical match for the zebrafish adra2a was found in the Sanger Institute's zebrafish genome database (Q90WY4); this gene was annotated on the basis of our GenBank entry for adra2a (AY048971). Differences from our adra2a sequence include the same I27V mutation in TM1 as above and another in the third intracellular (IC3) loop (D240E). The mutation I27V in both AW077216 and Q90WY4 is caused by the same nucleotide difference and is thus very likely to represent a true polymorphism, although the quality and reliability of the EST sequences is difficult to judge. However, no matches for the other genes identified here were found in the zebrafish genome database; searches with the cloned adra2 sequences identified only the adra2a sequence.

    RT-PCR

    As no information about the expression of the fourth 2-AR subtype genes is present from any species, an RT-PCR screening was performed. It shows that adra2db is expressed both in 48 to 50 hpf embryos and in adult zebrafish (fig. 1). For adra2da, no signal was present in the embryos and only a faint signal was seen in adult zebrafish (not shown), but this will require confirmation by in situ hybridizations.

    FIG. 1. RT-PCR reactions using adra2db-specific primers run on a 1.5% agarose gel. Lane 1: a 354-bp adra2db fragment amplified from genomic DNA. Lanes 2 and 4: the same 354-bp adra2db amplification product from RT-PCR using RNA isolated from 48 to 50 hpf embryos (lane 2) or adult zebrafish (lane 4) as templates. There are some smaller bands visible in both lanes, probably presenting prematurely terminated products. Lanes 3 and 5: same as above, but without RT to show that no genomic DNA contamination exists. The less than 100-bp band in lane 3 probably represents remnants of the template and primers. Lane 6: DNA size marker

    Sequence Analysis

    All five cloned zebrafish 2-AR genes are intronless, as are their mammalian orthologs, for the A, B, and C subtypes. At the protein level, the coding regions of the zebrafish and human 2-ARs differ in length (see the Supplementary Material online for the sequence alignment), mainly within the IC3, which connects TM5 to TM6, and within the N-terminal region. The cloned zebrafish 2-ARs share higher sequence identity (approximately 51% to 66%) with respect to their mammalian orthologs than to their mammalian paralogs (48% to 56%). The sequence identity among zebrafish 2-AR paralogs is about 52% to 55%, which is slightly higher than, for example, the identity among the human 2-AR paralogs (48% to 51%) and, by contrast, much lower than the identity between the zebrafish 2Da-AR and 2Db-AR (75%). The percentage sequence identities are provided in the Supplementary Material online.

    The most conserved regions among the 2-AR sequences correspond to the putative TM regions defined by similarity with the TM regions in bovine rhodopsin, whose crystal structure is known (Palczewski et al. 2000). The ligand-binding pocket is largely formed by the TM regions, and key residues implicated in binding interactions with adrenaline and noradrenaline (Nyronen et al. 2001) are all conserved among the 2-ARs. The two cysteine residues forming a disulfide bridge linking TM3 to the second extracellular (EC2) loop in the bovine rhodopsin crystal structure (Palczewski et al. 2000) are present in the 2-ARs. The GPCR consensus sequences DRY (TM3) and EKE (TM6), which are thought to form an ion pair that locks the rhodopsinlike GPCRs into an inactive state (e.g., Ballesteros et al. 2001), are also conserved. In the rhodopsin crystal structure, there are two palmitoylated cysteines located at the C-terminus, following a short cytoplasmic helix. In the human 2A-AR and 2B-AR, a single cysteine is present, but no cysteine is found in this region in the human 2C-AR (e.g., Kennedy and Limbird 1993). All of the zebrafish 2-ARs possess a cysteine at the equivalent position and therefore may be palmitoylated. With the exception of a site within the IC3 loop (E[S,T]4D), identified as a G-protein coupled receptor kinase (GRK) phosphorylation site in human 2A-AR (Eason, Moreira, and Liggett 1995; Jewell-Motz and Liggett 1995) and opossum 2C-AR (Deupree, Borgeson, and Bylund 2002), the regions outside of the TMs are more divergent. The equivalent site in the human 2C-AR (ESSAA) differs, is not phosphorylated, and its equivalent in 2B-ARs from eutherians could not be unambiguously identified, although the human 2B-AR is phosphorylated by GRKs. The zebrafish 2A-AR and 2C-AR possess the complete GRK phosphorylation site, whereas the site is present but incomplete in zebrafish 2B-ARs (ESMSSD), 2Da-AR (ESAASD), and 2Db-AR (ESSVSN). A similar site ((D/E)nxx(S/T)) is phosphorylated by Casein Kinase 1, and this mechanism has been proposed as an alternative pathway for GPCR phosphorylation (Tobin 2002).

    In contrast, conserved regions among the sequences of the fish and mammalian orthologs of a given subtype are mainly located within the IC1, IC2, EC1, and EC3 loops, near the C-terminus and within EC2 and IC3 adjacent to the TMs. The intracellular surface of 2-ARs interacts with G-proteins, where IC2 plays a role in selective coupling (Ostrowski et al. 1992), and parts of IC3 are required for Gs-coupling and may be required for Gi-coupling, too (Eason and Liggett 1996). Molecular fingerprints unique to each subtype, located 15 to 25 residues upstream from TM6, can be found within the N-terminal region of IC3 and correspond to the following regions in the zebrafish AR sequences: in 2A, residues 274 to 284 (Kx(K/R)xSQxKPG(D/E)); in 2B, residues 396 to 403 (ATx(K/R)GxxL); in 2C, residues 320 to 335 (RxSx(K/R)Sx(D/E)xFxSRR(K/R)R); in 2D (e.g., in 2Da) residues 280 to 282 (RFS) and residues 299 to 303 (RxSWA) were found common to fish and frog (no mammalian orthologs are known). In addition, a second messenger-regulated phosphorylation site for protein kinase C (PKC) in human 2A-AR has been identified at the C-terminus (Liang et al. 2002) of IC3, and sequences at these sites are conserved in fish 2A-ARs. Mammalian 2B-ARs possess a large stretch of glutamic acid residues within IC3 (discussed extensively in Madsen et al. 2002); a stretch of arginine and lysine residues are present in fish 2B-ARs, whereas a histidine repeat is present in 2C-AR from opossum and in lamprey 2-AR.

    The 2-AR sequences are most divergent along TM1 near the extracellular surface, near the N-terminus, within EC2, and within most of the long IC3 loop, showing little conservation even among fish and mammalian orthologs. The N-terminal region is N-glycosylated in the bovine rhodopsin crystal structure; N-glycosylation probably takes place in human 2A-ARs and 2C-ARs but not in 2B-AR, which lacks asparagine in this region (e.g., Keefer, Kennedy and Limbird 1994). All of the cloned zebrafish 2-ARs possess one or more asparagine residues within their N-terminal sequence, but their glycosylation states are unknown.

    The phylogenetic tree presented in figure 2 is based on an alignment of the available full-length 2-AR sequences, where only the regions unambiguously aligned among the 2-AR subtypes were considered. Only the tree calculated with the maximum-parsimony method is depicted here, but the Neighbor-Joining and maximum-likelihood methods gave very similar branching orders. The cuckoo wrasse "2F-AR" is a clear ortholog (high bootstrap support) of the zebrafish and mammalian 2C-AR subtypes. The putative 2-AR gene from goldfish (L09064) together with the zebrafish 2Da-AR and 2Db-AR form a separate, fourth 2-AR branch. The pufferfish 2-ARs also group together with their mammalian and zebrafish orthologs. The phylogenetic trees shown in figure 3 are based on alignments of translated partial nucleotide sequences retrieved from GenBank, whose subtypes were identified on the basis of a single tree constructed over all subtypes (data not shown). The available 2-AR sequences of the cluster comprising zebrafish, herring, pufferfish, toothcarp, and seahorse form two distinct groups within each of the four trees, representing the presence of duplicate subtypes, whereas the eel receptors do not always follow this pattern, although they are duplicated in the 2A-AR, 2C-AR, and 2D-AR subtypes, too. Subtypes A to D are present for frog and chameleon, but no duplicate 2-AR subtypes have been found. A single lamprey 2-AR sequence fragment has been identified (see Supplementary Material online), but this sequence is not clearly orthologous to any one of the four 2-AR subtypes.

    FIG. 2. Phylogenetic tree of 2-ARs produced using the maximum-parsimony method (see the Materials and Methods). The zebrafish 2-ARs are named according to the guidelines in http://zfin.org/zf_info/nomen.html, and the naming of the pufferfish 2-ARs is as they appear in the different databases. Replicate trees, 1,000, were used for bootstrap analysis. Bootstrap support: values shown next to nodes when less than 80%; closed circles indicate values greater than 80%. The tree was rooted by using bovine rhodopsin. This tree is not to scale

    FIG. 3. Ortholog-specific phylogenetic trees of the 2-AR subtypes including the nonmammalian fragments retrieved from GenBank (AL606538 to AL606586), produced using the maximum-parsimony method. The nomenclature of these gene fragments is that present in their GenBank entries. The nucleotide sequences were translated into protein sequences and the alignment was constructed as described in Materials and Methods. Replicate trees, 1,000, were used for the bootstrap analysis. Bootstrap support: values shown next to nodes when less than 80%; closed circles indicate values greater than 80%. The trees are rooted with the zebrafish 2-AR paralogs. Lengths of the horizontal branches are proportional to the amino acid sequence differences; the length of the scale bar is equal to a 20% difference. For every branch A, B, C, and D, the duplicates hypothetically common to ancestors of different species of fish are boxed. When two duplicates are present, the receptor and species name are underlined, with the exception of the eel receptor sequences, which show partially different branching pattern and lower bootstrap support compared with the other species with duplicates

    Southern Hybridizations

    Southern hybridizations on digested zebrafish genomic DNA under high and high-to-intermediate stringency revealed single bands of different sizes specific to adra2a, adra2b, and adra2c. Intermediate-to-low stringency conditions using each of these three zebrafish adra2 probes revealed one strong specific band corresponding to the probe used and two slightly weaker bands that corresponded to the other subtype genes, as defined by their sizes. The adra2da and adra2db probes recognized the duplicates even under high stringency conditions, revealing two bands. The identities of these bands were deduced on the basis of their intensity differences. The adra2da and adra2db bands were not visible under intermediate stringencies using the other three adra2 probes, most likely because of the somewhat lower sequence identity of these three sequences with the adra2d sequences in comparison with the percentage identities between the other three subtypes within the TM regions (see Supplementary Material online). Low stringency conditions revealed the same five bands corresponding to each of the five genes and an additional, weaker band (fig. 4). The identity of the remaining single band is currently unknown, but we tentatively conclude that no close second orthologs of the mammalian 2-AR subtypes exist in zebrafish (see Discussion).

    FIG. 4. Southern hybridization of HindIII-digested zebrafish genomic DNA hybridized with the adra2c-probe under low stringency conditions. The specific bands corresponding to different adra2 genes are indicated, as well as a barely visible 2.7-kb band of unknown identity

    Chromosomal Mapping of the Cloned Genes and Genome Comparisons

    We mapped the zebrafish adra2a gene to LG22, the adra2b gene to LG8, the adra2c gene to LG1, the adra2da gene to LG14, and the adra2db gene to LG21 (fig. 5) (zebrafish chromosomes are called linkage groups [LGs]). Portions of these chromosomes show conserved syntenies with mammalian chromosomes bearing the orthologs of adra2a, adra2b, and adra2c, as well as numerous other orthologous genes. Zebrafish adra2a is located very close to the genes cspg6 and mxi1 and likewise the human ortholog ADRA2A is close to CSPG6 and MXI1 on the human chromosome Hsa10. The adra2b gene is located close to mmp9 (positions 84.2 cM and 57.6 cM, respectively) on LG8. In addition, adra1a (an EST, AI641097), the only other adrenergic receptor currently mapped in the zebrafish, is also located on LG8, but the distance to adra2b and mmp9 is longer (position 16.7 cM). Mouse Adra2b and Mmp9 are located close to each other on the mouse chromosome Mmu2 (positions 71 cM and 96 cM, respectively). Adra1d is also located on Mmu2, between Adra2b and Mmp9 (position 74.9 cM). Human ADRA1D and MMP9 are located on Hsa20, whereas human ADRA2B is located on Hsa2p (Hsa2p and the HOX-bearing Hsa2q joined only recently in the human lineage to form the current chromosome Hsa2 [Larhammar, Lundin, and Hallbook 2002]). In addition, two ESTs (AI882753 and AI332283), probably orthologous to the genes DNMT3B and CC1.3 in human, respectively, are located to LG8 in zebrafish. This organization might suggest that the zebrafish EST classified as an 1A-AR actually is an ortholog of the human and mouse 1D-AR (see Discussion). For adra2c, much of the chromosome with many markers and genes, including mellar, tolloid, hmx1, msxb, lef1, fgp, and gpsn1 in zebrafish and their human orthologs ADRA2C, MTNR1A, TLL1, HMX1, MSX1, LEF1, FGP, and GPSN1, is conserved in a syntenic group between zebrafish LG1 and Hsa4. Compared with the large conserved synteny between human and zebrafish, however, the mouse genome has experienced a considerable amount of reorganization of these genes and chromosome segments among mouse chromosomes 3, 5, and 8. The adra2da and adra2db genes are located on separate linkage groups (LG14 and LG21, respectively), indicating that they are clearly separate genes and not alleles of the same gene. LG14 and Hsa5 display a syntenic group, including the genes egr1, glraz1, flt4, and kiaa0171 in the zebrafish and their human orthologs EGR1, GLRAZ1, FLT4, and KIAA0171. The zebrafish genes kiaa0731, kiaa0837, and kiaa0313 are located on LG21, whereas their human orthologs are on Hsa5 (KIAA0731 and KIAA0837) and on Hsa4 (KIAA0313). LG14 and LG21 also seem to bear duplicates comparable to adra2da and adra2db; the human ortholog of the duplicates msxa and msxd, the gene MSX2, is located on Hsa5. However, LG14 bears msxe, and LG1 bears the other duplicate, msx1, whereas their human orthlog, MSX1, is located on Hsa4. The human gene NKX3.2 on Hsa4 and its zebrafish ortholog nkx3.2 on LG14 also belong to this syntenic group. LG21 also shares some conserved syntenies with Hsa4 (in addition to the above mentioned kiaa0313, sap30, and dck on LG21 and their human orthologs SAP30 and DCK on Hsa4) as well as Hsa10 (cdk9 and sqt locate to LG21 and their human orthologs CDK9 and NODAL locate to Hsa10). ADRA1B and ADRB2 are also on Hsa5. In the mouse, there seems to have been rearrangements of these genes between Mmu11 (Adra1b, Glra1, and Flt4) and Mmu18 (Adrb2 and Egr1). Unfortunately, mapping data for the mouse orthologs was only available for a limited number of the above-mentioned genes.

    FIG. 5. Comparative syntenies of portions of zebrafish chromosomes (LG) containing the 2-AR subtype genes (adra2) relative to the human (Hsa) and mouse (Mmu) chromosomes. The gene order has been ignored to ease comparison of syntenies (genes on the same chromosome). The chart also contains the 1-AR genes (adra1) and ?-AR genes (adrb) for which mapping data were available and two additional ESTs: AI882753, mapped on the Ekker panel (retrieved from http://www.genetics.wustl.edu/fish_lab/frank/cgi-bin/fish/rhmaps/vector/vector08) and AI332283 mapped on the HS panel (supported by other funds, retrieved from http://zebrafish.stanford.edu). Nomenclature of genes is as in http://www.ncbi.nlm.nih.gov/genome/guide/human/ and in http://zfin.org/zf_info/nomen.html/. The zebrafish putative adra1a is highlighted in a white box (see text). ADRA1A and ADRB3 (in Hsa8) and ADRA2B (in Hsa2) are not shown as no syntenies were identified; the current location of these genes may have resulted from random events. Paralogous genes within species and orthologous genes among species are located on the same horizontal line. The fourth putative paralogy group on the right side of the panel contains genes encoding the mammalian 1B-AR and ?2-AR, and the zebrafish 2D-AR

    Discussion

    We report here five distinct zebrafish 2-AR genes. Three of these genes are counterparts of the mammalian 2-AR subtypes A, B, and C, thereby showing that they arose before the divergence of ray-finned fishes and tetrapods. Two of the zebrafish genes are closely related copies of a fourth 2-AR subtype. The protein sequences of zebrafish 2-ARs are 80% to 87% identical with their mammalian orthologs within the TM regions. This is approximately the same degree of sequence identity found between the fish and mammalian dopamine and 5-HT receptors (Cardinaud et al. 1997; Yamaguchi and Brenner 1997); it is higher than that observed for several peptide receptors such as the opioid and oxytocin receptors (Hausmann et al. 1995; Rodriguez et al. 2000). The zebrafish 2-AR paralogs share 74% to 81% sequence identity within their TM regions; thus, each zebrafish gene is more similar to a human gene than it is to another zebrafish gene, which suggests that the four 2-AR subtypes arose before the divergence of the ray-fin fish (actinopterygian) and lobe-fin fish (sarcopterygian) lineages. The sequence identity within the putative TM regions of the mammalian and fish 2-ARs is above 74%, and therefore we expect the TM bundle to be structurally conserved among the subtypes. Most of the key residues important for the binding of adrenaline and noradrenaline (Nyronen et al. 2001) are conserved across species and receptor subtypes, with the exception of the residue equivalent to cysteine 201 in human 2A-AR, where serine and threonine are also observed. It has recently been shown that the residue at position 201 is important for binding but less so for activation of human 2A-AR by catecholamines (Peltonen et al. 2003). Serine/cysteine variation at this position has also been shown to contribute to interspecies differences in antagonist binding affinity between human and mouse/rat 2A-ARs (Link et al. 1992).

    The lengths of the coding sequences vary considerably between the human and zebrafish orthologs, which can be explained for the most part by differences in the length of the third intracellular loop. For GPCRs in general, IC3 loop varies considerably among species orthologs, for example, between the human and pig Y5 NPY receptors (Wraith et al. 2000), as well as among the 2-AR subtypes. The human 2A and 2B subtypes are phosphorylated by GRKs, whereas the human 2C subtype is not. The phosphorylation site has been identified in human 2A-AR (Eason, Moreira, and Liggett 1995), but the phosphorylated serines/threonines remain to be identified in the human 2B-AR. In contrast to human, the fish 2B-ARs possess sequences compatible with a GRK phosphorylation site. At the equivalent location in the kangaroo and opossum 2B-ARs four serines/threonines of the consensus sequence are present, but the required acidic residue is absent; however, a repeat of acidic residues, equivalent to that found in eutherian mammals, is present about five residues upstream. In a naturally occurring variant of human 2B-AR, a deletion of three glutamate residues from the acidic residue repeat results in reduced phosphorylation and impaired desensitization of the receptor by GRKs (Small et al. 2001); this polymorphism is associated with a lowered basal metabolic rate in obese subjects (Heinonen et al. 1999) and an increased risk of acute coronary events, such as myocardial infarction (Snapir et al. 2001; 2003).

    We have identified distinct regions of the IC3-loop (approximately 15 to 25 residues upstream from the N-terminal end of TM6), unique to each subtype, that are conserved among fish and mammals, suggesting that they are functionally conserved and contribute to a distinct structural domain characteristic of each subtype. These sequence patterns may play a subtype-specific role within the 2-ARs related to G-protein specificity and binding or to the binding specificity of other intracellular effectors. In addition, these fingerprints provide strong external support that the classification of the 2-ARs into four subtypes is correct, as these fingerprints are not included in the sequence alignment leading to the phylogenetic tree shown in figure 2.

    The branching pattern in the phylogenetic tree (fig. 2) does not show obvious support for the hypothesis of two rounds of genome duplication before the divergence of actinopterygian and sarcopterygian lineages about 420 MYA (Amores et al. 1998), most probably because the first and second rounds of duplication took place very close to each other in time, leading to poor resolution of the root of the tree. It is likely that we do not have a sufficient number of phylogenetically informative sites to resolve this issue, and this is reflected in the relatively low bootstrap values at the root of the tree. The low bootstrap values at the root of the tree suggest that other evidence, for example chromosomal mapping data, is needed to resolve the early events in the phylogenetic history of the 2-ARs. The cuckoo wrasse "2F-AR" is clearly an ortholog of the zebrafish and mammalian 2C-ARs, and we suggest that it should be renamed as such. Likewise, the goldfish 2-AR should be named as 2D-AR. In naming the fourth subtype as 2D-AR (and the duplicate zebrafish genes as adra2da and adra2db) we have followed the IUPHAR Compendium of Receptor Characterization and Classification 2000 and nomenclature guidelines for naming the zebrafish genes. A potential confusion was initially caused by the inappropriate naming of the mouse/rat 2A-AR, and the misnomer should no longer be used.

    The 2D-AR subtype appears to be duplicated in zebrafish as well in pufferfish. The nature and timing of gene/genome duplications resulting in duplicated genes in early teleost evolution has been subject to a recent debate (see below). The fourth 2-AR subtype probably represents a result of the second duplication event, and the ancestral parent of the two zebrafish and pufferfish 2D-AR genes present in the last common ancestor of actinopterygians and sarcopterygians has apparently been lost in the mammalian lineage. There are other examples of this phenomenon, the retention of ancient duplicates in the fish lineage and their loss in the mammalian lineage. For example, there is an EVX gene adjacent to HOXA13 and HOXD13 but not at the corresponding position in the HOXB and HOXC clusters in mammals, suggesting that the pre–genome-expansion chromosome had an EVX gene at the 5' end of the HOX cluster, and after duplication events, it has been retained in some but not all of the chromosomes. In zebrafish, there is an additional evx gene adjacent to the hoxb cluster (Postlethwait et al. 1998, 2000). Thus, the last common ancestor of zebrafish and mammals must have had an EVX gene adjacent to the HOXB cluster. This has been retained in the zebrafish lineage but lost in the mammalian lineage. 2D-AR may share the functions of its ancestor (the common ancestor of 2D-AR and another 2-AR subtype) as suggested by the DDC (duplication-degeneration-complementation) hypothesis of the preservation of duplicated genes, according to which degenerative mutations in duplicate genes increase rather than reduce the probability of duplicate gene preservation, with the duplicate genes usually sharing the functions of their common ancestor rather than developing new functions (Force et al. 1999).

    As shown in figure 3, orthologs of the zebrafish 2D-ARs are present in higher, tetrapod vertebrates such as frog (Rana esculenta) and chameleon. This is consistent with the presence of this gene in the last common ancestor of actinopterygians and sarcopterygians and the loss of this gene in the mammalian lineage after the divergence of reptiles and mammals. Thus, it appears that the divergence of the actinopterygian and sarcopterygian lineages has occurred shortly after the two duplication events. In figure 3, the different duplicated fish 2-ARs are not always consistently named, and the eel 2-AR duplicates cannot always be classified as orthologs of the other duplicates. The eel 2-ARs display a somewhat conserved branching pattern with the group comprising seahorse, pufferfish, cuckoo wrasse, toothcarp, zebrafish, goldfish, and herring, although some eel 2-AR duplicates show low bootstrap values. The positioning of some 2-ARs is obviously not as expected (see for example the shark and chicken 2A-ARs in figure 3). Some of the duplications in different fish species may have taken place independently, such as with the "2A1" and "2A2" sequences in eel, or the number of informative sites in the available sequences may be too few to resolve this issue. Resolution of these questions would require much more sequence data from several different species. Additionally, mapping data and the comparison of the chromosomal arrangements of the genes would obviously be helpful. The identity of the single lamprey 2-AR is also impossible to solve with the currently available sequence data; it can either represent a descendant of a gene ancestral to all or represent just two of the four main 2-AR branches. Taylor et al. (2001) date the proposed third round of genome duplication to between 300 and 450 MYA on the basis of data from zebrafish, frog, chicken, human, and mouse. Critics of the hypothesis of a third round of tetraploidization early in teleost evolution point out the possibility of independent local duplications giving rise the abundance of fish genes (Robinson-Rechavi et al. 2001b) but also accept independent partial chromosomal duplications or whole-genome duplications as another possible explanation. Perhaps most importantly, they suggest that the analysis of several fish species and comparative linkage analysis can help to address this question (Robinson-Rechavi et al. 2001a). Our identification of four 2-AR subtype genes with one duplicated subtype in conserved chromosomal segments (see below) supports block duplications comparable to that of the HOX-clusters (Larhammar, Lundin, and Hallbook 2002). The complete octet of 2-ARs in the pufferfish genome is a particularly strong support for the one-to-four-to-eight duplication scheme for the number of paralogs after three rounds of tetraploidizations (Taylor et al. 2001).

    The adra2a gene is in a conserved synteny among zebrafish, human, and mouse. The adra2c gene is in a conserved synteny between human and zebrafish. In the mouse, a considerable amount of reorganization of the mouse orthologs of the syntenic group formed by adra2c and other genes on LG1 and their orthologs on human chromosome 4 seem to have taken place among mouse chromosomes 3, 5, and 8. The adra2b gene is in a conserved synteny between zebrafish and mouse (fig. 5). In mouse and cat, orthologs of human chromosomes 20 and 2p are syntenic, implying that this is the ancestral mammalian condition. In zebrafish, too, orthologs of Hsa20 and Hsa2p are syntenic on several chromosomes, especially the duplicated chromosome pair LG17 and LG20. LG8 has one locus from Hsa20, MMP9, and one from Hsa2p, ADRA2B. The zebrafish configuration with mmp9 and adra2b probably represents the ancestral organization, although there are many loci from several other human chromosomes on LG8. Furthermore, the two additional zebrafish ESTs mapped on LG8 and their possible human and mouse orthologs on Hsa20 and Mmu2 support this scheme (see figure 5). The situation with the EST possibly representing the zebrafish adra1a gene is not clear. Because it is based on a short EST sequence, its classification may not be correct. It is not located very close to adra2b and mmp9, and even if it would really represent adra1d, its association might still be random. The genes adra2da and adra2db are located on LG14 and LG21, respectively. Both LG14 and LG21 show conserved syntenies with Hsa5 and Mmu11 and Mmu18, but also with Hsa4 and Hsa10, so our grouping of portions of LG14, LG21, Hsa5, Mmu11, and Mmu18 into the fourth paralogy group is speculative. However, on the basis of the phylogenetic analyses, we conclude that this arrangement is the most plausible one.

    In conclusion, the localization of the cloned zebrafish 2A-AR, 2B-AR, 2C-AR, 2Da-AR, and 2Db-AR genes seems to reflect conserved syntenies between zebrafish and human and/or mouse, which corroborates the sequence-based subtype classification presented in this paper. The molecular phylogenetic analysis and the comparative mapping results are consistent with the hypothesis that the initial events in generating different adrenergic receptor subtypes were local duplications, giving rise to ancestors of the 1-AR, 2-AR, and ?-AR genes. All known 2-AR subtype genes (with the possible exception of the pufferfish "2D2-AR" gene) and most of the ?-AR subtype genes are intronless, as are most vertebrate GPCRs, which has been suggested to reflect retrotransposition as the mechanism of duplication (Gentles and Karlin 1999). However, as the ancestral 1-AR, 2-AR, and ?-AR genes seem to have been located on the same chromosomal segment, it would seem more plausible that they arose from an ancestral adrenoceptor gene by local tandem duplications. After two rounds of block or large-scale (chromosomal, or even genome) duplication, the subtypes evolved by random genetic drift, accompanied by purifying selection, leading to the divergence of sequences among subtypes or species (fig. 6). If the duplications took place as part of two tetraploidization events, one should expect a fourth 2-AR gene copy, as has been found in zebrafish and pufferfish. Orthologs of 2D-AR also seem to be present in at least some other species of fish and even in some tetrapod vertebrates, but so far they have not been found in mammals. However, it is quite common that duplicate genes are either totally lost or present as nonfunctional pseudogenes, as shown by many other gene families, for instance, the HOX-gene cluster (Sharman and Holland 1998). It is very likely that no such gene will be found in the well-characterized human, mouse, and rat genomes, but the existence of such a gene would require considerable reevaluation of the current concepts concerning the functions and physiological roles of the 2-ARs.

    FIG. 6. Hypothetical scheme for the duplication of the ancestral AR-gene (ADR) during vertebrate evolution. Local duplications (I) of the ancestral AR-gene resulted in three copies of the original gene, the ancestors of the current main classes of ARs (gene nomenclature as in figure 5). The first round of genome duplication (II) results in two copies of the original chromosome segment. A second genome duplication (III) results in four copies of the original segment. This probably represents the arrangement in a common vertebrate ancestor of the bony fishes and tetrapods. Subsequent chromosomal rearrangements, probable inactivation of the fourth subtypes (marked with a question mark for ADRB and ADRA1), along with speciation and sequence divergence resulted in the current subtypes present in vertebrates. The putative third round of duplication in the fish lineage is not shown. This scheme is in accordance with the phylogeny and current arrangement of the 2-AR subfamily in mammals and zebrafish presented in figures 2, 3, and 5

    Southern hybridizations using gradually lowered stringencies suggest that no close additional orthologs (co-orthologs) of the three mammalian 2-AR subtypes exist in zebrafish; band(s) representing these orthologs should have been visible under intermediate-to-low stringency conditions, when each subtype-specific probe also recognized other paralogs. The extra band visible under low stringency conditions might represent a faster evolving 2-AR copy (second ortholog) or another GPCR. The latter possibility is supported by the fact that despite several screenings with probes capable of recognizing a distantly related receptor (the μ-opioid receptor), we have not found any duplicates of the 2A-AR, 2B-AR, or 2C-AR gene in zebrafish, as predicted by the hypothesis of an ancient, fish-specific third round of polyploidization; in contrast, pufferfish has a complete octet of 2-ARs. Fish species other than pufferfish, including zebrafish, seem to have experienced differential gene loss after the generation of eight 2-AR genes by three rounds of duplication in their common ancestor. Completion of the zebrafish genome project should give a definitive answer as to the number of zebrafish 2-ARs but only after the genes identified by the genome sequencing project have been thoroughly analyzed. To date, only one of the genes reported here is available on the zebrafish genome project Web site. This single adra2a gene was annotated on the basis of our GenBank entry, indicating the importance of making a thorough analysis of the genes identified by random sequencing.

    It is interesting to note that the duplication scheme for the adrenergic receptors is very similar to that of the neuropeptide Y receptors located on the same chromosomal segments in mammals (Wraith et al. 2000). Although orthologous NPY receptors have not yet been identified in zebrafish (Larhammar et al. 2001), recent observations in a cartilaginous fish, the spiny dogfish (a shark), confirm that the gene duplications took place before the gnasthostome radiation. Furthermore, the shark NPY receptor study showed that the anatomical distribution of receptor mRNA (as detected by RT-PCR) seems to be quite different between sharks and mammals (Salaneck et al. 2003). ESTs representing the zebrafish 2A-AR and 2B-AR, the cDNA representing 2C-AR, and the RT-PCR results for 2Db-AR (and possibly also for 2Da-AR) show only that these receptor genes are transcribed, with no correlation of the expression to anatomical structures or possibility to deduce their functions. Our recent unpublished results from HPLC indicate that the concentration of noradrenaline in the adult zebrafish brain is substantial and comparable to that in rat brain: 4.53 ± 0.97 nmol/mg tissue (mean ± SD, n = 15). Our preliminary binding assays indicate that the 2-AR antagonist radioligand [ethyl-3H]RS-79948-197 displays binding in membranes prepared from zebrafish brain homogenates comparable with that observed for the cloned zebrafish receptors. A receptor density (Bmax) of 475 fmol/mg protein and an affinity constant (Kd) of 0.1 nM was obtained in a single experiment on 21 pooled brains. The Kd values determined for the different zebrafish 2-ARs (subtypes A, B, C, Da, and Db), expressed separately in CHO cells, ranged from 0.1 to 0.7 nM (Ruuskanen and Scheinin, unpublished data). Work is in progress to investigate the tissue distribution patterns of the zebrafish 2-AR subtypes.

    In this work, we have shown that the events generating the diverse 2-AR subtypes took place before the divergence of ancestors of teleost fish and tetrapods, and we have identified a fourth, duplicated 2-AR subtype that gives further support for the ancient fish-specific genome duplication hypothesis and opens new insights into the study of the adrenergic receptors. These results indicate evolutionary pressure to maintain a larger set of 2-AR subtypes in the fish genomes than is known for mammals and suggest distinct and important biological functions for each of these receptor subtypes.

    Acknowledgements

    Maija Ivaska is thanked for skillful technical assistance. The study was funded by grants from the Academy of Finland and Technology Development Center of Finland to M.S.J and M.S., a grant from the Swedish Natural Science Research Council (NFR) to D.L., and grants R01 RR10715 and P01HD22486 from NIH to J.H.P.

    Literature Cited

    Akimenko, M. A. 1995. Preparation of genomic DNA for Southern analysis. P. 9.11 in M. Westerfield, ed. The zebrafish book. Guide for the laboratory use of zebrafish (Danio rerio). 3rd edition. University of Oregon Press, Eugene.

    Amores, A., A. Force, and Y. L. Yan, et al. (13 co-authors). 1998. Zebrafish hox clusters and vertebrate genome evolution. Science 282:1711-1714.

    Ballesteros, J. A., A. D. Jensen, G. Liapakis, S. G. Rasmussen, L. Shi, U. Gether, and J. A. Javitch. 2001. Activation of the beta 2-adrenergic receptor involves disruption of an ionic lock between the cytoplasmic ends of transmembrane segments 3 and 6. J. Biol. Chem. 276:29171-29177.

    Bargmann, C. I. 1998. Neurobiology of the Caenorhabditis elegans genome. Science 282:2028-2033.

    Barrallo, A., R. Gonzalez-Sarmiento, F. Alvar, and R. E. Rodriguez. 2000. ZFOR2, a new opioid receptor-like gene from the teleost zebrafish (Danio rerio). Brain. Res. Mol. Brain. Res. 84:1-6.

    Blaxall, H. S., D. R. Cerutis, N. A. Hass, L. J. Iversen, and D. B. Bylund. 1994. Cloning and expression of the alpha 2C-adrenergic receptor from the OK cell line. Mol. Pharmacol. 45:176-181.

    Blaxall, H. S., D. A. Heck, and D. B. Bylund. 1993. Molecular determinants of the alpha-2D adrenergic receptor subtype. Life Sci. 53:L255-L259.

    Bockaert, J., and J. P. Pin. 1999. Molecular tinkering of G protein-coupled receptors: an evolutionary success. EMBO J. 18:1723-1729.

    Bylund, D. B., H. S. Blaxall, L. J. Iversen, M. G. Caron, R. J. Lefkowitz, and J. W. Lomasney. 1992. Pharmacological characteristics of alpha 2-adrenergic receptors: comparison of pharmacologically defined subtypes with subtypes identified by molecular cloning. Mol. Pharmacol. 42:1-5.

    Cardinaud, B., K. S. Sugamori, S. Coudouel, J. D. Vincent, H. B. Niznik, and P. Vernier. 1997. Early emergence of three dopamine D1 receptor subtypes in vertebrates: molecular phylogenetic, pharmacological, and functional criteria defining D1A, D1B, and D1C receptors in European eel Anguilla anguilla. J. Biol. Chem. 272:2778-2787.

    Chatwin, H. M., J. E. Rudling, D. Patel, V. Reale, and P. D. Evans. 2003. Site-directed mutagenesis studies on the Drosophila octopamine/tyramine receptor. Insect Biochem. Mol. Biol. 33:173-184.

    Deupree, J. D., C. D. Borgeson, and D. B. Bylund. 2002. Down-regulation of the alpha-2C adrenergic receptor: involvement of a serine/threonine motif in the third cytoplasmic loop. BMC Pharmacol. 2:9.

    Eason, M. G., and S. B. Liggett. 1996. Chimeric mutagenesis of putative G-protein coupling domains of the alpha2A-adrenergic receptor. Localization of two redundant and fully competent Gi coupling domains. J. Biol. Chem. 271:12826-12832.

    Eason, M. G., S. P. Moreira, and S. B. Liggett. 1995. Four consecutive serines in the third intracellular loop are the sites for beta-adrenergic receptor kinase-mediated phosphorylation and desensitization of the alpha 2A-adrenergic receptor. J. Biol. Chem. 270:4681-4688.

    Felsenstein, J. 1993. PHYLIP (phylogeny inference package). Version 3.5c. Distributed by the author. Department of Genetics, University of Washington, Seattle.

    Flugge, G., M. van Kampen, H. Meyer, and E. Fuchs. 2003. Alpha2A and alpha2C-adrenoceptor regulation in the brain: alpha2A changes persist after chronic stress. Eur. J. Neurosci. 17:917-928.

    Force, A., M. Lynch, F. B. Pickett, A. Amores, Y. L. Yan, and J. Postlethwait. 1999. Preservation of duplicate genes by complementary, degenerative mutations. Genetics 151:1531-1545.

    Gentles, A. J., and S. Karlin. 1999. Why are human G-protein-coupled receptors predominantly intronless? Trends Genet. 15:47-49.

    Gerhardt, C. C., R. A. Bakker, G. J. Piek, R. J. Planta, E. Vreugdenhil, J. E. Leysen, and H. Van Heerikhuizen. 1997. Molecular cloning and pharmacological characterization of a molluscan octopamine receptor. Mol. Pharmacol. 51:293-300.

    Guo, S., S. W. Wilson, S. Cooke, A. B. Chitnis, W. Driever, and A. Rosenthal. 1999. Mutations in the zebrafish unmask shared regulatory pathways controlling the development of catecholaminergic neurons. Dev. Biol. 208:473-487.

    Guyer, C. A., D. A. Horstman, A. L. Wilson, J. D. Clark, E. J. J. Cragoe, and L. E. Limbird. 1990. Cloning, sequencing, and expression of the gene encoding the porcine alpha 2-adrenergic receptor. Allosteric modulation by Na+, H+, and amiloride analogs. J. Biol. Chem. 265:17307-17317.

    Hausmann, H., W. Meyerhof, H. Zwiers, K. Lederis, and D. Richter. 1995. Teleost isotocin receptor: structure, functional expression, mRNA distribution and phylogeny. FEBS Lett. 370:227-230.

    Heinonen, P., M. Koulu, U. Pesonen, M. K. Karvonen, A. Rissanen, M. Laakso, R. Valve, M. Uusitupa, and M. Scheinin. 1999. Identification of a three-amino acid deletion in the alpha2B-adrenergic receptor that is associated with reduced basal metabolic rate in obese subjects. J. Clin. Endocrinol. Metab. 84:2429-2433.

    Hill, C. A., A. N. Fox, R. J. Pitts, L. B. Kent, P. L. Tan, M. A. Chrystal, A. Cravchik, F. H. Collins, H. M. Robertson, and L. J. Zwiebel. 2002. G protein-coupled receptors in Anopheles gambiae. Science 298:176-178.

    Holland, P. W., F. J. Garcia, N. A. Williams, and A. Sidow. 1994. Gene duplications and the origins of vertebrate development. Dev. Suppl. 125–133.

    Jewell-Motz, E. A., and S. B. Liggett. 1995. An acidic motif within the third intracellular loop of the alpha2C2 adrenergic receptor is required for agonist-promoted phosphorylation and desensitization. Biochemistry 34:11946-11953.

    Johnson, M. S., and J. P. Overington. 1993. A structural basis for sequence comparisons: an evaluation of scoring methodologies. J. Mol. Biol. 233:716-738.

    Katugampola, S., and A. Davenport. 2003. Emerging roles for orphan G-protein-coupled receptors in the cardiovascular system. Trends Pharmacol. Sci. 24:30-35.

    Keefer, J. R., M. E. Kennedy, and L. E. Limbird. 1994. Unique structural features important for stabilization versus polarization of the alpha 2A-adrenergic receptor on the basolateral membrane of Madin-Darby canine kidney cells. J. Biol. Chem. 269:16425-16432.

    Kelly, P. D., F. Chu, and I. G. Woods, et al. (15 co-authors). 2000. Genetic linkage mapping of zebrafish genes and ESTs. Genome Res. 10:558-567.

    Kennedy, M. E., and L. E. Limbird. 1993. Mutations of the alpha 2A-adrenergic receptor that eliminate detectable palmitoylation do not perturb receptor-G-protein coupling. J. Biol. Chem. 268:8003-8011.

    Kobilka, B. K., H. Matsui, T. S. Kobilka, F. T. Yang, U. Francke, M. G. Caron, R. J. Lefkowitz, and J. W. Regan. 1987. Cloning, sequencing, and expression of the gene coding for the human platelet alpha 2-adrenergic receptor. Science 238:650-656.

    Lanier, S. M., S. Downing, E. Duzic, and C. J. Homcy. 1991. Isolation of rat genomic clones encoding subtypes of the alpha 2-adrenergic receptor: identification of a unique receptor subtype. J. Biol. Chem. 266:10470-10478.

    Larhammar, D., L. G. Lundin, and F. Hallbook. 2002. The human Hox-bearing chromosome regions did arise by block or chromosome (or even genome) duplications. Genome Res. 12:1910-1920.

    Larhammar, D., A. Wraith, M. M. Berglund, S. K. Holmberg, and I. Lundell. 2001. Origins of the many NPY-family receptors in mammals. Peptides 22:295-307.

    Liang, M., M. G. Eason, C. T. Theiss, and S. B. Liggett. 2002. Phosphorylation of Ser360 in the third intracellular loop of the alpha2A-adrenoceptor during protein kinase C-mediated desensitization. Eur. J. Pharmacol. 437:41-46.

    Link, R., D. Daunt, G. Barsh, A. Chruscinski, and B. Kobilka. 1992. Cloning of two mouse genes encoding alpha 2-adrenergic receptor subtypes and identification of a single amino acid in the mouse alpha 2-C10 homolog responsible for an interspecies variation in antagonist binding. Mol. Pharmacol. 42:16-27.

    Lomasney, J. W., W. Lorenz, L. F. Allen, K. King, J. W. Regan, F. T. Yang, M. G. Caron, and R. J. Lefkowitz. 1990. Expansion of the alpha 2-adrenergic receptor family: cloning and characterization of a human alpha 2-adrenergic receptor subtype, the gene for which is located on chromosome 2. Proc. Natl. Acad. Sci. USA 87:5094-5098.

    MacDonald, E., B. K. Kobilka, and M. Scheinin. 1997. Gene targeting–homing in on alpha 2-adrenoceptor-subtype function. Trends Pharmacol. Sci. 18:211-219.

    Madsen, O., M. Scally, C. J. Douady, D. J. Kao, R. W. DeBry, R. Adkins, H. M. Amrine, M. J. Stanhope, W. W. de Jong, and M. S. Springer. 2001. Parallel adaptive radiations in two major clades of placental mammals. Nature 409:610-614.

    Madsen, O., D. Willemsen, B. M. Ursing, U. Arnason, and W. W. de Jong. 2002. Molecular evolution of the mammalian alpha 2B adrenergic receptor. Mol. Biol. Evol. 19:2150-2160.

    Meyer, H., M. Palchaudhuri, M. Scheinin, and G. Flugge. 2000. Regulation of alpha(2A)-adrenoceptor expression by chronic stress in neurons of the brain stem. Brain Res. 880:147-158.

    Miyata, T., and H. Suga. 2001. Divergence pattern of animal gene families and relationship with the Cambrian explosion. Bioessays 23:1018-1027.

    Naruse, K., S. Fukamachi, and H. Mitani, et al. (20 co-authors). 2000. A detailed linkage map of medaka, Oryzias latipes: comparative genomics and genome evolution. Genetics 154:1773-1784.

    Nyronen, T., M. Pihlavisto, and J. M. Peltonen, et al. (14 co-authors). 2001. Molecular mechanism for agonist-promoted alpha(2A)-adrenoceptor activation by norepinephrine and epinephrine. Mol. Pharmacol. 59:1343-1354.

    Ostrowski, J., M. A. Kjelsberg, M. G. Caron, and R. J. Lefkowitz. 1992. Mutagenesis of the beta 2-adrenergic receptor: how structure elucidates function. Annu. Rev. Pharmacol. Toxicol. 32:167-183.

    Palacios, J. M., B. F. O'Dowd, S. Cotecchia, M. Hnatowich, M. G. Caron, and R. J. Lefkowitz. 1989. Adrenergic receptor homologies in vertebrate and invertebrate species examined by DNA hybridization. Life Sci. 44:2057-2065.

    Palczewski, K., T. Kumasaka, and T. Hori, et al. (12 co-authors). 2000. Crystal structure of rhodopsin: A G protein-coupled receptor. Science 289:739-745.

    Pebusque, M. J., F. Coulier, D. Birnbaum, and P. Pontarotti. 1998. Ancient large-scale genome duplications: phylogenetic and linkage analyses shed light on chordate genome evolution. Mol. Biol. Evol. 15:1145-1159.

    Peltonen, J. M., T. Nyronen, and S. Wurster, et al. (11 co-authors). 2003. Molecular mechanisms of ligand-receptor interactions in transmembrane domain V of the alpha(2A)-adrenoceptor. Br. J. Pharmacol. 140:347-358.

    Postlethwait, J. H., Y. L. Yan, and M. A. Gates, et al. (29 co-authors). 1998. Vertebrate genome evolution and the zebrafish gene map. Nat. Genet. 18:345-349.

    Postlethwait, J. H., I. G. Woods, P. Ngo-Hazelett, Y. L. Yan, P. D. Kelly, F. Chu, H. Huang, A. Hill-Force, and W. S. Talbot. 2000. Zebrafish comparative genomics and the origins of vertebrate chromosomes. Genome Res. 10:1890-1902.

    Regan, J. W., T. S. Kobilka, F. T. Yang, M. G. Caron, R. J. Lefkowitz, and B. K. Kobilka. 1988. Cloning and expression of a human kidney cDNA for an alpha 2-adrenergic receptor subtype. Proc. Natl. Acad. Sci. USA 85:6301-6305.

    Rex, E., and R. W. Komuniecki. 2002. Characterization of a tyramine receptor from Caenorhabditis elegans. J. Neurochem. 82:1352-1359.

    Robinson-Rechavi, M., O. Marchand, H. Escriva, and V. Laudet. 2001a. Re: Revisiting recent challenges to the ancient fish-specific genome duplication hypothesis. Curr. Biol. 11:R:1007-R1008.

    Robinson-Rechavi, M., O. Marchand, H. Escriva, and V. Laudet. 2001b. An ancestral whole-genome duplication may not have been responsible for the abundance of duplicated fish genes. Curr. Biol. 11:R458-R459.

    Rodriguez, R. E., A. Barrallo, F. Garcia-Malvar, I. J. McFadyen, R. Gonzalez-Sarmiento, and J. R. Traynor. 2000. Characterization of ZFOR1, a putative delta-opioid receptor from the teleost zebrafish (Danio rerio). Neurosci. Lett. 288:207-210.

    Salaneck, E., D. H. Ardell, E. T. Larson, and D. Larhammar. 2003. Three neuropeptide Y receptor genes in the spiny dogfish, Squalus acanthias, support en bloc duplications in early vertebrate evolution. Mol. Biol. Evol. 20:1271-1280.

    Sharman, A. C., and P. W. Holland. 1998. Estimation of Hox gene cluster number in lampreys. Int. J. Dev. Biol. 42:617-620.

    Small, K. M., K. M. Brown, S. L. Forbes, and S. B. Liggett. 2001. Polymorphic deletion of three intracellular acidic residues of the alpha 2B-adrenergic receptor decreases G protein-coupled receptor kinase-mediated phosphorylation and desensitization. J. Biol. Chem. 276:4917-4922.

    Snapir, A., P. Heinonen, and T. P. Tuomainen, et al. (14 co-authors). 2001. An insertion/deletion polymorphism in the alpha2B-adrenergic receptor gene is a novel genetic risk factor for acute coronary events. J. Am. Coll. Cardiol. 37:1516-1522.

    Snapir, A., J. Mikkelsson, M. Perola, A. Penttila, M. Scheinin, and P. J. Karhunen. 2003. Variation in the alpha2B-adrenoceptor gene as a risk factor for prehospital fatal myocardial infarction and sudden cardiac death. J. Am. Coll. Cardiol. 41:190-194.

    Spring, J. 2003. Major transitions in evolution by genome fusions: from prokaryotes toeukaryotes, metazoans, bilaterians and vertebrates. J. Struct. Funct. Genomics 3:19-25.

    Springer, M. S., G. C. Cleven, O. Madsen, W. W. de-Jong, V. G. Waddell, H. M. Amrine, and M. J. Stanhope. 1997. Endemic African mammals shake the phylogenetic tree. Nature 388:61-64.

    Springer, M. S., E. C. Teeling, O. Madsen, M. J. Stanhope, and W. W. de Jong. 2001. Integrated fossil and molecular data reconstruct bat echolocation. Proc. Natl. Acad. Sci. USA 98:6241-6246.

    Stanhope, M. J., O. Madsen, V. G. Waddell, G. C. Cleven, W. W. de-Jong, and M. S. Springer. 1998. Highly congruent molecular support for a diverse superordinal clade of endemic African mammals. Mol. Phylogenet. Evol. 9:501-508.

    Svensson, S. P., T. J. Bailey, D. J. Pepperl, N. Grundstrom, S. Ala-Uotila, M. Scheinin, J. O. Karlsson, and J. W. Regan. 1993. Cloning and expression of a fish alpha 2-adrenoceptor. Br. J. Pharmacol. 110:54-60.

    Taylor, J. S., Y. Van de Peer, I. Braasch, and A. Meyer. 2001. Comparative genomics provides evidence for an ancient genome duplication event in fish. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 356:1661-1679.

    Taylor, J. S., Y. Van de Peer, and A. Meyer. 2001. Revisiting recent challenges to the ancient fish-specific genome duplication hypothesis. Curr. Biol. 11:R1005-R1008.

    Teeling, E. C., O. Madsen, R. A. Van den Bussche, W. W. de Jong, M. J. Stanhope, and M. S. Springer. 2002. Microbat paraphyly and the convergent evolution of a key innovation in Old World rhinolophoid microbats. Proc. Natl. Acad. Sci. USA 99:1431-1436.

    Tobin, A. B. 2002. Are we beta-ARKing up the wrong tree? Casein kinase 1 alpha provides an additional pathway for GPCR phosphorylation. Trends Pharmacol. Sci. 23:337-343.

    Venkataraman, V., T. Duda, and R. K. Sharma. 1997. The bovine alpha 2D-adrenergic receptor gene: structure, expression in retina, and pharmacological characterization of the encoded receptor. Mol. Cell Biochem. 177:113-123.

    Venter, J. C., M. D. Adams, and E. W. Myers, et al. (274 co-authors). 2001. The sequence of the human genome. Science 291:1304-1351.

    Vernier, P., B. Cardinaud, O. Valdenaire, H. Philippe, and J. D. Vincent. 1995. An evolutionary view of drug-receptor interaction: the bioamine receptor family. Trends Pharmacol. Sci. 16:375-381.

    Woods, I. G., P. D. Kelly, F. Chu, P. Ngo-Hazelett, Y. L. Yan, H. Huang, J. H. Postlethwait, and W. S. Talbot. 2000. A comparative map of the zebrafish genome. Genome Res. 10:1903-1914.

    Wraith, A., A. Tornsten, P. Chardon, I. Harbitz, B. P. Chowdhary, L. Andersson, L. G. Lundin, and D. Larhammar. 2000. Evolution of the neuropeptide Y receptor family: gene and chromosome duplications deduced from the cloning and mapping of the five receptor subtype genes in pig. Genome Res. 10:302-310.

    Yamaguchi, F., and S. Brenner. 1997. Molecular cloning of 5-hydroxytryptamine (5-HT) type 1 receptor genes from the Japanese puffer fish, Fugu rubripes. Gene 191:219-223.(Jori O. Ruuskanen*,, Henr)