A cDNA clone encoding the soluble guanylyl cyclase α2 subunit was isolated from medaka fish (Oryzias latipes) and designated as OlGCS-α2. The OlGCS-α2 cDNA was 3192 bp in length and the open reading frame (ORF) encodes a protein of 805 amino acids. The deduced amino acid sequence has high similarity to that of the mammalian α2 subunit gene except for the N-terminal regulatory domain. The C-terminal 5 amino acids, “RETSL”, which have been reported to interact with the post synaptic density protein (PSD)-95 were conserved. An RNase protection assay with adult fish organs showed that OlGCS-α2 was expressed mainly in the brain and testis. The complete nucleotide sequence (about 41 kbp) of the OlGCS-α2 genomic DNA clone isolated from a medaka fish BAC library indicated that the OlGCS-α2 gene consisted of 9 exons and 8 introns. The 5′-flanking region and larger introns, such as introns 1, 4, and 7, contained the several fragments conserved in the nucleotide sequences of Rex6 (non-long terminal repeat retrotransposon), MHC class I genomic region, and OlGC1, the medaka fish homolog of the mammalian guanylyl cyclase B gene. Linkage analysis on the medaka fish chromosome demonstrated that the OlGCS-α2 gene was mapped to LG13; this mapping position was different from those for the OlGCS-α1 and OlGCS-β1 genes (LG1).
INTRODUCTION
Soluble GC, a nitric oxide (NO) and carbon monoxide (CO)-sensitive guanylyl cyclase (GC), is an enzyme that catalyzes the conversion of GTP to cGMP through binding of these gaseous ligands to the heme in the enzyme (Gerzer et al., 1981). cGMP thus synthesized by the soluble GC activates cGMP-dependent protein kinases (cGKs), phosphodiesterases (PDEs), and cyclic-nucleotide gated (CNG) cation channels. Through the activation of these downstream effectors, the NO/cGMP signaling pathway plays important roles in various physiological phenomena, e.g., smooth muscle relaxation, platelet aggregation, and neural development (Schmidt and Walker, 1994; Kusakabe and Suzuki, 2000; Lucas et al., 2000).
Soluble GC is a heterodimeric enzyme consisting of α and β subunits (Garbers et al., 1994). In mammals, two iso-forms of each subunit have been identified (α1 and α2, β1 and β2). The α1/β1 heterodimer was purified first from the mammalian lung (Gerzer et al., 1981), and subsequently, the cDNA clone for each subunit was isolated and sequenced (Koesling et al., 1988). The α2 and β2 subunit cDNA clone were isolated by homology screening from a human fetal brain and rat kidney cDNA library, respectively (Yuen et al., 1990; Harteneck et al., 1991). The α1 or α2 subunit was demonstrated to form an active heterodimeric enzyme (α1/β1 or α2/β1) with the β1 subunit when both the subunit cDNAs were co-expressed in Sf9 cells (Russwurm et al., 1998), while the β2 subunit has been reported to be able to form an active enzyme in the absence of a second subunit (α1 or α2) (Koglin et al., 2001).
It has been demonstrated that the α2 subunit mRNA is expressed in the human brain, uterus, and placenta (Budworth et al., 1999), and the α2 subunit protein was detected in the placenta, where the β1 subunit protein was also detected (Russwurm et al., 1998). Moreover, it has been demonstrated the interaction of the C-terminal peptide of soluble GC α2 subunit with the post synaptic density-95 protein (PSD-95), which is a synaptic scaffold protein, suggesting that the α2/β1 heterodimer can be recruited to the membrane (Russwurm et al., 2001). Several studies have reported that both the α1/β1 heterodimeric soluble GC and the endothelial NO synthase (eNOS) are translocated in a caveolae of the lung endothelial cells, and Hsp 90 acts as a scaffold protein between these proteins (Zabel et al., 2002; Nedvetsky et al., 2002). Together, these findings indicate the potential importance of the α2/β1 heterodimeric soluble GC, which shows enzyme kinetics and a tissue distribution similar to those of the α1/β1 heterodimeric soluble GC. Recently, we demonstrated that the α2 subunit mRNA was expressed during embryogenesis of the medaka fish Oryzias latipes and suggested that the α2/β1 heterodimeric enzyme plays an important role in the early development of the eye (Harumi et al., 2003; Yamamoto et al., 2003). In this study, we report the structures of the cDNA and genomic DNA clones encoding the medaka fish soluble GC α2 sub-unit gene (OlGCS-α2) and assess their tissue distribution by means of an RNase protection assay using various adult medaka fish organs.
MATERIALS AND METHODS
Animals
Mature adults of the orange-red variety of the medaka fish O. latipes were maintained as described previously (Yamagami et al., 2001). Mature male individuals of the O. latipes Hd-rR inbred strain (Hyodo-Taguchi and Sakaizumi, 1993) were fixed in ethanol and used for isolation of the genomic DNA.
Preparation of RNA and isolation of an OlGCS-α2 cDNA frag-ment by RT-PCR
Total RNA was prepared from the adult brain and kidney of the orange-red variety of the medaka fish O. latipes using TRIZOL™ reagent (Invitrogen, Carlsbad, CA, USA). Poly (A)+ RNA was isolated using Oligotex-dT30<Super> (Roche, Mannheim, Germany), according to the manufacturer's protocol. Three μg of the poly (A)+ RNA was reverse-transcribed with Superscript II (Invitrogen) in 50 μl scale. The degenerate oligonucleotide primers were designed and synthesized based on the amino acid sequences conserved in all reported soluble GC subunit proteins (sense-KGQMI: 5′-CCCGCGGAATTCAGCTTMRIGGICARATGRTI-3′; antisense-MPRYCLF: 5′-GAATTCTCGAGGATCCRAAIARRCARTAICIIGGCAT-3′). The first PCR amplification was carried out with the first strand cDNA as a template and performed for 30 cycles under the following reaction conditions: 94°C for 1 min (denaturation), 50°C for 1 min (annealing), 72°C for 45 sec (elongation), and an additional elongation reaction for 5 min at 72°C. Then, the second PCR was performed using the nested primers synthesized based on the amino acid sequences conserved in mammalian soluble GC α2 (a2-5′-1: 5′-GCNAARGCNCARGAYGG-3′ for the amino acid sequence AKAQDG) and the antisense-MPRYCLF primer that was used in the first PCR. The PCR products were separated by electrophoresis with a 1.5% SeaKem GTG agarose gel (BMA, Rockland, ME, USA) and purified using MinElute Gel Extraction Kit (QIAGEN, Hilden, Germany). The cDNA fragments were subcloned into the plasmid vector pBluescript II KS (Stratagene, La Jolla, CA, USA) and sequenced.
5′- and 3′-Rapid Amplification of cDNA Ends (5′- and 3′-RACE)
To obtain the full-length sequence of the OlGCS-α2 cDNA, the 5′-portion of the cDNA was amplified by the 5′-RACE method (Frohman et al., 1988) with a 5′-RACE System for Rapid Amplification of cDNA Ends, ver. 2.0 (Invitrogen). Briefly, 1 μg of total RNA isolated from the adult brain was reverse-transcribed with several gene-specific antisense primers (5′A: 5′-ACTTGCGAGCAGGCACTGGC-3′ [cDNA nucleotide no. 2023-2004]; 5′D: 5′-ATCAGCGAGCAGGATCCGGC-3′ [1733–1713]; 5′G: 5′-AAGATCCGCCTGGCAACAGC-3′[1130-1111]; 5′J: 5′-CACGGCTCGAAGAACTCGC-3′ [876–858]). The cDNA was tailed with dCTP using terminal deoxynucleotidyl transferase and then amplified with the Abridged Anchor Primer and gene-specific primers (5′B: 5′-TTCTGTGCCACATCACCCGG-3′ [1988-1969]; 5′E: 5′-TTCTTTGGTGCCGGCCTGCG-3′ [1677-1658]; 5′H: 5′-ACAGCTCTGATCAAGCCTGG-3′ [1115–1096]; 5′K: 5′-CAGAGACCAAAGAACTCTTCGC-3′ [842-821]). Nested PCR was performed with the Abridged Anchor Primer and nested gene-specific primers (5′C: 5′-GGTGGGTCCGCTCTAATGTG-3′ [1915-1896]; 5′F: 5′-GGTCCGGATGGTAAAGGGGG-3′ [1645–1625]; 5′I: 5′-GCAGGGTTGAATCAGTGCAG-3′ [1073–1054]; 5′L: 5′-GTTCTCGCAGTTCACAAAACGG-3′ [814–792]). The 3′-portion of the cDNA was amplified by the 3′-RACE method (Frohman et al., 1988) using the 3′-Full RACE Core Set (TaKaRa, Otsu, Japan). Total RNA (1 μg) of the medaka fish brain was reverse-transcribed with an Oligo dT-3′ sites Adaptor Primer and gene-specific 3′Z primer: 5′-ATACTGTGTGGCTGGAGGAC-3′ [2187–2206]. The second PCR was carried out with the 3′ site Adaptor Primer and nested gene-specific 3′Y primer: 5′-AATTCACACAGGCTCGGTGC-3′ [2322–2341]. The RACE products were subcloned into pBluescript II KS vector and sequenced.
Determination of the transcription start site of the OlGCS-α2 gene by the oligo-capping 5′-RACE method
The transcription start site of the OlGCS-α2 gene was determined by the oligo-capping 5′-RACE method (Maruyama and Sugano, 1994) using a First Choice™ RLM-RACE kit (Ambion, Austin, TX, USA). All steps were carried out according to the manufacturer's protocol. Adaptor-ligated RNA from the total RNA (10 μg) of the medaka fish brain was reverse-transcribed with the gene-specific 5′N primer: 5′-AGCTGCGTCCGTTCCAGAGG-3′ [454–435]. The first PCR was carried out with the 5′-RACE Outer Primer and the gene-specific 5′O primer: 5′-GAGGAGCGCTCTTTGGGAGG-3′ [438–419]. The conditions were 30 sec at 96°C, 30 sec at 60°C, and 30 sec at 72°C for 35 cycles, followed by elongation at 5 min at 72°C. The nested PCR was performed with the 5′-RACE Inner Primer and nested gene-specific 5′P primer: 5′-CCGAGCTACTGAATGACTCG-3′ [316–297], and the PCR program was the same as that for the first PCR. The RACE products were subcloned into pBluescript II KS vector and sequenced.
Molecular phylogenetic analysis
The partial amino acid sequence (residues 528 to 766) of OlGCS-α2 was compared with those of the corresponding part of known fish and mammalian soluble GC subunit isoforms using the Clustal W program (Thompson et al., 1994) and the sequence editor SeqPub (Gilbert, Indiana University). An unrooted phylogenetic tree was constructed using the aligned sequences by means of the neighbor-joining algorithms (Saitou and Nei, 1987) in the PROTRAS program of PHYLIP version 3.572 (Felsenstein, 1989) and Clustal W program (Thompson et al., 1994). For neighbor joining analysis, the evolutionary distance was estimated using Kimura's empirical method for protein distances (Kimura, 1983).
GenBank/EMBL/DDBJ accession numbers for the sequences used for comparison are as follows: human GCS-α1 ( Y15723), rat GCS-α1 ( M57405), FrGCS-α1 ( AB062171), OlGCS-α1 ( AB000849), human GCS-α2 ( X63282), rat GCS-α2 ( AF109963), human GCS-β1 ( X66533), rat GCS-β1 ( M22562), FrGCS-β1 ( AB062172), OlGCS-β1 ( AB000850), human GCS-β2 ( NM_004129), rat GCS-β2 ( AB058888), and OlGC1 ( AB004921).
RNase protection assay
The cDNA fragment of 333 bp, 412 bp, or 216 bp containing the 3′-UTR region of OlGCS-α1 (2200–2532), OlGCS-β1 (1897–2308), or OlGCS-α2 (2694–2909) was subcloned into pBluescript II KS vector for preparation of the probe. After digestion with EcoRI, a cRNA probe was synthesized using T3 RNA polymerase with ATP, CTP, GTP, and [α-32P] UTP and a DIG RNA Labeling kit (Roche) according to the manufacturer's protocol. The synthesized probe (1×105 cpm) was treated with RNase-free DNase I (Roche), then extracted by phenol and purified using CHROMA SPIN-30 columns (CLONTECH, Palo Alto, CA, USA). The purified cRNA probe was applied to the pool of total RNA (10 μg) extracted from various adult medaka fish organs (brain, eye, gill, heart, gall bladder, spleen, kidney, testis, ovary, liver, and intestine). The mixture was ethanol-precipitated and dissolved in a hybridization buffer containing 80% formamide, 40 mM Pipes (pH 6.4), 400 mM NaCl, and 1 mM EDTA, followed by incubation to anneal each other overnight at 50°C. Single stranded RNA was treated with RNase A for 30 min at 30°C in the solution containing 300 mM NaCl, 10 mM Tris-HCl (pH 7.4), and 5 mM EDTA. The protected fragment was treated with 125 μg/ml Proteinase K and 0.5% SDS, and then phenol/chloroform-extracted, ethanol-precipitated, and electrophoresed on a 6% polyacrylamide gel containing 7 M urea. The gel was dried and analyzed using a FUJIX Bio-Imaging Analyzer BAS2000 (Fuji Photo Film, Tokyo, Japan).
Northern hybridization
Poly (A)+ RNA (7.5 μg) from various organs of medaka fish was separated on 1% agarose gel containing 6.7% formaldehyde. The RNA was transferred to a nylon membrane, Hybond-H+ (Amersham Pharmacia Biotech, Little Chalfont Bucks, UK) with 10xSSPE as a transferring solution. A cDNA fragment (nucleotides 226–1645) of OlGCS-α2 was labeled with [α-32P] dCTP using the Random Primer DNA Labeling kit Version 2 (TaKaRa) and used as a probe. The blot was pre-hybridized in 50% formamide, 5xSSPE, 5xDenhardt's solution, 0.5% SDS, and 100 μg/ml denatured herring sperm DNA at 42°C for 1 hr. The radioactive probe was added to the pre-hybridization buffer and incubated overnight at 42°C. The radioactive signals were visualized using a FUJIX Bio-Imaging Analyzer BAS2000.
Isolation of genomic DNA clones for OlGCS-α2 from a medaka fish bacterial artificial chromosome (BAC) library
A high-density replica (HDR) membrane of an O. latipes Hd-rR inbred strain genomic BAC library (Asakawa et al., 1997; Matsuda et al., 2001) was used for screening of the OlGCS-α2 gene. The treatments of the pre-hybridization membrane were performed as described previously (Yamagami et al., 2001). To isolate the OlGCS-α2 gene, hybridization was carried out using a probe constructed by PCR with the OlGCS-α2 cDNA as a template and the following primers: precap: 5′-TGCATCCCCCTTTACCATCC-3′; a2tail: 5′-AAATCCAAAGCTCAGCACCC-3′. Positive BAC clones were detected using CDP-star detection reagent (Amersham Pharmacia Biotech) according to the manufacturer's instructions. A QIAGEN plasmid maxi kit (QIAGEN) was used for BAC DNA isolation from the bacterial culture. BAC DNA was digested with EcoRI and HindIII, and then subjected to Southern hybridization with the same probe described above to confirm the isolation of positive clones. Following the Southern hybridization, to check whether or not the clones contained the full-length of the OlGCS-α2 gene, PCR was performed with the following primer pairs: a2test: 5′-CTGCACTGATTCAACCCTGC-3′ and 5′G: 5′-AAGATCCGCCTGGCAACAGC-3′; UTRUP: 5′-TCAGACCGTGTTACAAAGGC-3′; and a2tail: 5′-AAATCCAAAGCTCAGCACCC-3′. The conditions were as follows: 30 cycles at 96°C for 30 sec, 61°C for 30 sec, and 72°C for 1 min, and an additional incubation at 72°C for 5 min.
Genomic Southern hybridization
A membrane being blotted with the restriction enzyme-treated genomic DNA of an individual of the O. latipes Hd-rR inbred strain was prepared as described previously (Yamagami et al., 2001). The membrane was pre-hybridized for at least 1 hr at 42°C in a solution containing 50% formamide, 5xSSPE, 5xDenhardt's solution, 0.5% SDS, and 100 μg/ml denatured herring sperm DNA. A 592 bp cDNA fragment of OlGCS-α2 (1054–1645) was labeled with [α-32P] dCTP using the Random Primer DNA Labeling kit version 2 (TaKaRa) and was used as a probe. The radioactive probe was added to the prehybridization solution, followed by incubation overnight at 42°C. The membrane was washed three times with 2xSSC/0.1% SDS at 50°C for 15 min. Imaging of the radioactive signals was performed with a FUJIX Bio-Imaging Analyzer BAS2000 (Fuji Photo Film, Tokyo, Japan).
Linkage analysis of the OlGCS-α1, -α2, and -β1 subunit genes
Assignment of the loci encoding OlGCS-α1, OlGCS-α2, OlGCS-β1 to each linkage group was carried out by the method described previously (Naruse et al., 2000). The primers and restriction enzymes used were as follows: LR-RT for OlGCS-α1, 5′-GTAAAAGAAATGTGGGGA-3′; LF-2 for OlGCS-α1, 5′-TTATTGATGTCTGACAGCCTA-3′; Mse I for OlGCS-α1; 5′d for OlGCS-α2, 5′-TAGGAACATGGTTCCAATGCTG-3′; 3′-Y for OlGCS-α2, 5′-AATTCACACAGGCTCGGTGC-3′; Hae III for OlGCS-α2; s-b1 for OlGCS-β1, 5′-AGTACAAGCTGACCCAAG-3′; s-b5 for OlGCS-β1, 5′-TCTGTCAGGATGTCAAAG-3′; Hae III for OlGCS-β1.
Other methods
The nucleotide sequence of cDNA and genomic DNA fragments was determined by the dideoxy chain termination method (Sanger et al., 1977) with an ABI PRISM™ 3100 Genetic Analyzer (Applied Biosystems, Foster city, CA, USA). Sequence data was analyzed with GENETYX-MAC/version 7.2.0 (Software Development, Tokyo, Japan). The homology search was performed at the Web site NCBI BLAST ( http://www.ncbi.nlm.hih.gov/BLAST/).
RESULTS
Characterization of OlGCS-α2 cDNA
To obtain the cDNA fragment of the medaka fish soluble GC α2 subunit gene, RT-PCR was performed with three degenerate primers (antisense-MPRYCLF, sense-KGQMI, and AKAQDG) synthesized based on the conserved amino acid sequences among mammalian soluble GCs. A homology search of a 526 bp cDNA fragment obtained from the medaka fish brain and kidney samples revealed that it was a partial nucleotide sequence of the cDNA of a medaka homolog of the mammalian soluble GC α2 subunit gene and was designated as OlGCS-α2. By performing repeated 5′-and 3′-RACE, the full-length cDNA of OlGCS-α2 was obtained. It was 3192 bp in length, which size was in agreement with the result of Northern blot analysis using the adult brain RNA (data not shown). The OlGCS-α2 cDNA consisted of the 2418 bp-open reading frame (ORF) and the 507 bp-3′-untranslated region (UTR).
As shown in Fig. 1, the oligo-capping 5′-RACE with the adult brain RNA demonstrated that there were nine distinct transcription start sites at the 13 to 267 nucleotides upstream of the translation start site “ATG”, and the nucleotide “C” at 267 bp upstream of the first methionine was designated as “+1”. There were GC-rich nucleotides and no typical TATA box around these transcription start sites, which findings were in good agreement with those for the gene having the TATA-less promoter (Smale, 1997).
Fig. 1
Complete nucleotide sequence of the OlGCS-α2 cDNA. Arrows indicate the transcriptional start sites. Vertical lines indicate exonintron boundaries. The numbers at the end of each line indicate the number of nucleotides or amino acids. The degenerate primers used in RTPCR were synthesized based on the boxed amino acids.
Comparison and phylogenetic analysis of the amino acid sequences of OlGCS-α2 with those of other soluble GC subunits
The deduced amino acid sequence (805 residues) of the OlGCS-α2 cDNA was aligned with those of the mammalian soluble GC α2 cDNAs (Fig. 2A). The OlGCS-α2 consisted of a regulatory domain (residues 1 to 401), a central domain (residues 402 to 539), and a catalytic domain (residues 540 to 766). The catalytic domain of OlGCS-α2 was 79.5% identical to that of rat soluble GC α2 and 59.4% identical to that of OlGCS-α1 (Fig. 2B). The central domain was 69.6% and 56.7% identical to those of rat soluble GC α2 and OlGCS-α1, respectively (Fig. 2B). However, the N-terminal regulatory region of OlGCS-α2 had several amino acid insertions (7 to 15 residues) compared with that of mammalian soluble GC α2, and its similarity to those of mammalian soluble GC α2 and OlGC-α1 was low at 36.3% and 41.2%, respectively (Fig. 2A). Moreover, the C-terminal 5 amino acids “RETSL” of OlGCS-α2 were almost the same as those of the motif that has been demonstrated to be the site for interaction with rat brain PSD-95 (Russwurm et al., 2001).
Fig. 2
(A) Alignment of amino acid sequence of OlGCS-α2 with those of mammalian soluble GC α2 subunits. The degenerate primers used in RT-PCR were synthesized based on the amino acids designated by thick arrows at the top of the sequence. Boxed amino acids indicate sites that may interact with PSD-95 (Russwurm et al., 2001). An asterisk indicates an amino acid that is identical among the three proteins, and a pair of dots indicates two amino acids that are identical and one amino acid that has a similar nature among the three proteins. A single dot indicates that the two proteins have an identical amino acid and one amino acid with a different nature. Gaps in the sequence are indicated by dashes. Domains are designated as suggested by Stone and Marletta (1995). Exon-intron boundaries of the OlGCS-α2 gene are designated by a V-shaped symbol with the number of exons given beside it. (B) Sequence identity and similarity among the predicted amino acid sequences of three domains (N-terminal, central, and catalytic domain) of known soluble GC subunits. (C) Molecular phylogenetic tree of OlGCS-α2 and the other soluble GC subunits. The amino acid sequences of various soluble GC subunits were subjected to phylogenetic analysis and the amino acid sequence of OlGC1 was used as an outgroup. The numbers indicate the bootstrapping value. Sources and their accession numbers are described in the MATERIALS AND METHODS. Abbreviations: Ol, medaka fish Oryzias latipes; Fr, Fugu fish Fugu rubripes.
As shown in Fig. 2C, the molecular phylogenetic analysis using the amino acid sequence of the catalytic domain of OlGCS-α2 (residues 528 to 766) and those of the corresponding domain of other soluble GC subunits (α1, α2, β1, and β2) indicated that OlGCS-α2 belonged to the α2 group to which the rat and human soluble GC α2 subunits belonged.
Expression of OlGCS-α2 in various medaka fish adult organs assayed by the RNase protection method
The organ distribution of the three subunit mRNAs (OlGCS-α1, OlGCS-α2, and OlGCS-β1) was examined by the RNase protection method using the total RNA from various medaka fish adult organs. The results indicated that all mRNAs were expressed mainly in the brain and testis, although a weak signal due to the OlGCS-α2 mRNA was detected in the eye, gall bladder, spleen, ovary, and intestine, where OlGCS-β1 was also expressed (Fig. 3).
Characterization of genomic DNA clones for OlGCS-α2
An O. latipes Hd-rR strain genomic BAC library was screened to isolate the OlGCS-α2 genomic clone with the partial cDNA fragment of OlGCS-α2 (nucleotides 1543–2921), resulting in detection of 24 positive clones from 18432 clones of the medaka fish BAC library. To confirm the positivity of these clones, we carried out Southern hybridization and PCR amplification experiments using the probe described above. The results indicated that 15 out of 24 clones contained the OlGCS-α2 gene and 12 out of the 15 clones possessed the 5′-UTR of OlGCS-α2, and, subsequently, that 8 out of these 12 clones contained the 3′-UTR of OlGCS-α2. We chose the clone 156J21 for use in later experiments. By sequencing this BAC clone, we finally determined the complete nucleotide sequence of 41 kbp for the OlGCS-α2 gene (Fig. 4). Furthermore, we determined 6 kbp nucleotide sequences of the 5′-flanking region of the OlGCS-α2 gene (data not shown). The sequence upstream of the transcription start sites contained GC-rich sequences and no canonical TATA box. As shown in Fig. 4 and Table 1, the OlGCS-α2 gene consisted of 9 exons which was the same number of exons in the OlGCS-α1 gene (Mikami et al., 1999), and the GT-AG rule was conserved for all splice sites (Table 1). Introns 1 (11,056 bp), 4 (11,211 bp), and 7 (9,817 bp) were especially larger than the others. In the 5′-flanking region (nucleotides from −1186 to −773) of the OlGCS-α2 gene, we found the nucleotide sequences that were conserved in Rex 6, a non-long terminal repeat (LTR) retrotransposon (Volff et al., 2001). The nucleotide sequences in intron 1 (nucleotides 3737–3914) and intron 4 (nucleotides 21498–21658 and 21614–21658) contained the sequences conserved in the MHC class I genomic region (Matsuo et al., 2002). Furthermore, in the nucleotide sequences in intron 7 (nucleotides 30279–30490 and 36894–37100), we found the sequences conserved in other MHC-related genes, immunoproteasome β subunit gene (PSMB9) and the transporter associated with antigen presentation gene (ABCB3) (accession no. AB073378). Most of these elements were also conserved in the nucleotide sequences of intron 6 and 15 of the OlGC1 gene (a medaka fish homolog of atrial natriuretic peptide receptor type B) (Fig. 4 and Table 2). In particular, the 5′-flanking region of the OlGCS-α2 gene contained a 1.2 kbp-sequence found in the OlGC-1 gene (Fig. 4A).
Fig. 4
Characteristic regions found in the 5′-flanking region (A) and intron 1 (B), and intron 7 (C) of the OlGCS-α2 gene. Diagrams at the top and bottom schematically show the genomic structure of OlGCS-α2 and OlGC1; the scale is indicated below the genomic structure of OlGC1. Black boxes with numbers indicate exons and horizontal lines denote introns. The six regions conserved in the two introns are indicated by boxes with various patterns, and identical patterns are used to denote regions having almost identical nucleotide sequences. The nucleotide sequences in the fragment R5′-3 (A) and R7-1 (C) were highly conserved with that in Rex 6 and PSMB9/ABCB3, respectively.
Table 1
Exon/Intron organization of OlGCS-α 2
Table 2
Characteristic regions of OlGCS-α 2 containing the fragments conserved with OlGC-1
Genomic Southern analysis and Linkage analysis
Genomic Southern hybridization was performed using a 591 bp cDNA fragment as a probe and revealed a single positive band in each of three lanes (Fig. 5). The size of the positive bands was consistent with those of the DNA fragments obtained from the digestion of genomic clones by the restriction enzymes, suggesting that the medaka fish genome contains a single copy of the OlGCS-α2 gene. We also carried out linkage mapping of the OlGCS-α1, OlGCS-α2, and OlGCS-β1 genes on the medaka fish chromosome, and demonstrated that the OlGCS-α1 and OlGCS-β1 genes were mapped to LG 1, while the OlGCS-α2 gene was mappedto LG 13.
DISCUSSION
In the present study, we demonstrated that the medaka fish O. latipes possessed the soluble guanylyl cyclase α2 subunit gene (designated as OlGCS-α2), and that this gene was expressed in the organs where the OlGC-β1 gene— whose translation product was the essential counterpart of that of the OlGCS-α2 gene—was expressed, although its chromosomal localization was different. It has been reported that the soluble GC α2 subunit gene is expressed in the human fetal brain (Harteneck et al., 1991), and that the soluble GC α2 subunit forms an active heterodimer (α2/β1) with the β1 subunit, which shows enzymatic characteristics similar to those of the α1/β1 heterodimer (Russwurm et al., 1998). To date, there has been no report showing the existence of the soluble GC α2 subunit in non-mammalian animals. In this study, by RT-PCR using cDNA prepared from total RNAs of the adult medaka fish brain and kidney and primers synthesized based on the amino acid sequences conserved among all soluble GC subunits, we obtained a cDNA fragment having high similarity to that of mammalian soluble GC α2 subunits (Fig. 2). The catalytic domain at the C-terminal of OlGCS-α2 was highly conserved among various soluble GC subunits, while the regulatory domain at the N-terminus containing the insertion of 7–15 amino acids was much less similar than those of mammalian soluble GC α2 subunits (Fig. 2A, 2B). In a previous study, we demonstrated that the regulatory region of OlGCS-α1 had low similarity to those of the soluble GC α1 subunits of other species (Mikami et al., 1998). Therefore, we presume that these regions were not particularly important for the function of the enzyme, and thus many mutations which might have occurred during its molecular evolution were accumulated in the regulatory region of the α subunit genes. The C-terminal five amino acid residues, RETSL, of the soluble GC α2 sub-unit were reported to interact with the PDZ domain of rat brain PSD-95, which has been proposed to form a signaling complex with other membrane proteins, including neuronal nitric oxide synthase (nNOS) and the N-methyl-D-aspartate (NMDA) receptor (Russwurm et al., 2001). As described above, OlGCS-α2 also possessed the same residues in the C-terminal region, suggesting that they play similar roles in medaka fish.
In the RNase protection assay of the OlGCS-α2 tran-scripts, the signals due to the OlGCS-α2 transcripts were mainly detected in the brain, eye, testis, and ovary (Fig. 3). Recently, we demonstrated using in situ hybridization that the OlGCS-α2 gene was expressed in the medaka fish embryonic brain and retina and that the expression in the embryonic retina became weaker with as the development proceeded (Harumi et al., 2003; Yamamoto et al., 2003). The expression of the OlGCS-α2 gene in the gonad is in good agreement with a recent report on the expression of the soluble GC subunit genes in mice (Mergia et al., 2003).
The OlGCS-α2 gene consisted of 9 exons and 8 introns (Fig. 4); some introns, such as introns 1, 4, and 7, were very large (Table 1) and contained many fragments conserved in introns 6 and 15 of a medaka fish homolog of the atrial natriuretic receptor gene (OlGC-1) (Takeda and Suzuki, 1999). The OlGC-1 gene contained several repeated nucleotide sequences conserved in Rex 6, a non-LTR retrotransposon (Volff et al., 2001), and the MHC class I genomic region (Matsuo et al., 2002). These results suggest that several genetic recombinations via transposable elements between the OlGCS-α2 and OlGC-1 genes occurred in the process of the OlGCS-α2 gene evolution, and this idea might be extensible to the mechanism of generation of the diverse numbers of membrane and soluble GC isoforms in vertebrates over a long period of time.
Linkage mapping of three soluble GC subunit genes (OlGCS-α1, OlGCS-α2, and OlGCS-β1) demonstrated that the OlGCS-α2 gene was located in LG 13, which was different from the location (LG 1) of the OlGCS-α1 and OlGCS-β1genes. In our previous report, the OlGCS-α1 and OlGCS-β1 genes were aligned tandemly in the medaka fish genome, separated by a 1 kbp-spacer sequence (Mikami et al., 1999). The α1/β1 and α2/β1 heterodimers are considered to be the only active soluble GCs in vertebrates. Therefore, it is rational to expect that the α1 and α2 subunits should be in competition for the association with the β1 subunit to form an active enzyme. Moreover, the 5′-flanking regions of the OlGCS-α1 and OlGCS-β1 genes were shown to mutually influence each other's promoter activity in a study measuring the promoter activity in mammalian cultured cells and medaka fish embryonic cells (Yamamoto and Suzuki, 2002). On the other hand, we demonstrated that the OlGCS-α2 and OlGCS-β1 genes, but not the OlGCS-α1 gene, were co-expressed in the embryonic retina of the medaka fish O. latipes (Harumi et al., 2003). Taking these results together, we presume that in some organs the transcription and/or translation of the OlGCS-α2 and OlGCS-β1 genes are coordinated, and in the other organs the transcription and/or translation of the OlGCS-α1 and OlGCS-β1 genes are coordinated. In the former case, the transcription and/or translation of the OlGCS-α1 gene should be repressed and in the latter case the transcription and/or translation of the OlGCS-α2 gene should be repressed. In either case, the expression of the soluble GC subunit genes could be regulated at the transcriptional and/or translational level.
To date, the α2 subunit of soluble GC has not attracted the attention of many investigators, probably due to the dearth of available information on its genomic structure, the organ distribution of the transcripts, and its function, relative to the many studies on the α1 and β1 subunits. However, our present and recent studies demonstrating that the wide distribution of the α2 subunit mRNA and inhibition of translation of the OlGCS-α2 gene by means of an antisense oligonucleotide caused severe defects in medaka fish embryos should contribute to a deeper understanding of the unsolved but important biological roles of the α2 subunit (Yamamoto et al., 2003). The differential functions between the α1/β1 and α2/β1 heterodimers in the NO/cGMP signaling pathway remained to be solved.
Acknowledgments
The authors would like to thank Ms. H. Kuboshita for culturing of medaka fishes and the staff members of the Center for the Advanced Sciences and Technology, Hokkaido University for the use of their laboratory facilities. This work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, Sports, and Culture of Japan (no. 11236202), the Japan Society for the Promotion of Science (no. 13010976), and the National Project on Protein Structural and Functional Analyses.
