- Research article
Ancient intron insertion sites and palindromic genomic duplication evolutionally shapes an elementally functioning membrane protein family
BMC Evolutionary Biologyvolume 7, Article number: 143 (2007)
In spite of the recent accumulation of genomic data, the evolutionary pathway in the individual genes of present-day living taxa is still elusive for most genes. Among ion channels, inward K+ rectifier (IRK) channels are the fundamental and well-defined protein group. We analyzed the genomic structures of this group and compared them among a phylogenetically wide range with our sequenced Halocynthia roretzi, a tunicate, IRK genomic genes.
A total of 131 IRK genomic genes were analyzed. The phylogenic trees of amino acid sequences revealed a clear diversification of deuterostomic IRKs from protostomic IRKs and suggested that the tunicate IRKs are possibly representatives of the descendants of ancestor forms of three major groups of IRKs in the vertebrate. However, the exon-intron structures of the tunicate IRK genomes showed considerable similarities to those of Caenorhabditis. In the vertebrate clade, the members in each major group increased at least four times those in the tunicate by various types of global gene duplication. The generation of some major groups was inferred to be due to anti-tandem (palindromic) duplication in early history. The intron insertion points greatly decreased during the evolution of the vertebrates, remaining as a unique conservation of an intron insertion site in the portion of protein-protein interaction within the coding regions of all vertebrate G-protein-activated IRK genes.
From the genomic survey of a family of IRK genes, it was suggested that the ancient intron insertion sites and the unique palindromic genomic duplication evolutionally shaped this membrane protein family.
Although a vast amount of data on the genomic structures of genes of major phyla in animal, plant, and microbial kingdoms has been accumulated as a result of many recent successful projects of whole genome shotgun DNA sequencing of biologically and medically important taxa [1–7], the evolutional pathway and physiological significance of individual genome gene structures, such as the conservation of intron insertions, duplication of genes, and development of gene regulatory sequences of the present-day living taxa are still elusive and unclear for most genes. Recent large population studies derived from the established genome databases have promoted our understanding of intron significance, favoring the exon theory or the intron early theory, though many points remain to be discussed and clarified [8–12]. Similarly, we know that gene duplication by the tandem repeat or chromosomal duplication must be the major evolutional power to drive simple to complex living systems at least in the vertebrate clade [13–17]. However, for individual genes the exact evidence is still in short availability because of the gene divergence of the whole genome chromosomes in the present-day organisms on the evolutionary pathway from the ancestor organisms, though the recent excellent whole genome studies on the Saccharomycetes have greatly advanced our understanding about these matters [18–20]. Further, although the eukaryotic genomic complexity may be initiated by the non-adaptive fixation of genetic drift resulting from a reduction of the effective population size , at least in the secondary development, it is plausible to correlate the evolution of the genomic structure to the adaptation of organisms to new environmental niches, such as, the functional evolution of the proteins as the gene products . However, little is known regarding the mechanisms except the proposal that the borders of domains in the protein are represented by the exon boundaries, thus being a proposal of physiological support for the exon theory [23–25].
Ion channels are one of the major and important protein groups functioning for regulating the intra- and extracellular ionic environments and for the signal transduction of intracellular and intercellular systems . Among ion channels, the inward K+ rectifier (IRK) channels comprise the most fundamental and simple protein group, having two transmembrane segments and one pore-forming region [27, 28]. The physiological functions are well studied in both protostomia and deuterostomia, and almost all members of genes have been sequenced at least in the mammals, such as human, mouse, and rat, by cDNA cloning or by genome projects [26, 28].
Recently, we have sequenced three Halocynthia roretzi IRK genomic clones, of which cDNA clones have been known [29, 30], being TuIRKA, TuGIRKAa, and TuGIRKB clones. Furthermore, the urochordate, Ciona intestinalis, genome projects  have allowed a genomic survey of tunicate IRK channel genes for molecular evolutionary analysis in comparison with the genes of both deuterostomal and protostomal clades. This has occurred because the urochordate genes have been known to locate at the branching point for the early vertebrate divergence within the deuterostomal clades, as shown in the recent exhaustive comparative study of ion channel genes viewed from the annotation of the Ciona unochordate genes .
In the present study, we attempted to analyze the genomic structures of a group of proteins, which are definitely defined functionally, and to compare them among a phylogenetically wide range, including protostomal to deuterostomal clades. We also aimed to elucidate the functional and evolutional significance of intronic insertion, intronic preservation, gene duplication on the chromosomes, and the relationship between the genome structure and the expressed protein structure or physiological function of the organism.
Genomic sequences of tunicate inward K+ rectifier genes
We determined genomic sequences of three inward K+ rectifier genes, TuIRKA, TuGIRKAa, and TuGIRKB, cloned from a genomic library of a tunicate, Halocynthia roretzi [see Additional file 1]. With previously reported respective cDNA sequences [29, 30] [see Additional file 2], their exons and introns in their coding regions were precisely determined [see Additional files 1 and 2]. These sequences covered entire coding regions, including all introns, 5' upper stream regions, and most of the 3' lower stream regions (Fig. 1 and [see Additional file 3]). All intron exon junctions, which existed in the coding regions, revealed the mammalian-conserved splicing characteristic sequences, such as the GT of splice donor sites at the initiation of introns and the AG of splice acceptor sites at the 3' terminal of respective introns (Fig. 2 and [see Additional file 4]). The average exon length was about 100 bp and was considerably short in comparison with those found in mammalian Kcnj genes. The average intron length was about 500 bp; thus these genomic sequences revealed considerably large numbers of intron insertions, from 8 to 11, within a rather short whole coding length of about 1000 bp of IRK genes [see Additional file 1]. Although the insertions were apparently evenly distributed among the coding regions, all three IRK genomes revealed characteristic intron insertions in the functionally important and thereby the strongly conserved sequence portions of the membrane-spanning regions and the ion pore regions or limes of those regions (Fig. 1).
The phylogenetic tree and a comparison of Halocynthia IRK AA sequences with those of other animal IRK genes
Thanks to many recent whole genome projects, there are many numbers of IRK genomic sequences, which are determined over their whole genome regions. Among them, as the first step, the genomic sequences of the same tunicate Ciona IRK genes, human IRK genes, Caenorhabditis elegans IRK genes, Drosophila melanogaster, and Anopheles gambiae genes were compared with the sequences of Halocynthia roretzi IRK genes. With the amino acid (AA) sequences inferred from the JGI Ciona genome database, from the NCBI GenBank and Genome database, or from the Ensembl database, the AA sequences of all IRK proteins from the seven taxa were aligned by using the ClustalX1.83 program, and their phylogenetic tree was made by the Neighbor Joining Method in the Mega3 v3.1 program (Fig. 3). As the outgroup, the seven bacterial IRK-like proteins were chosen, including KirBAC1.1, of which the molecular structure was recently determined by X-rays . Although some branches were not significant according to Bootstrap values, i.e., less than 50%, the tree revealed the following three points. First, the three IRK proteins in Halocynthia roretzi revealed corresponding close homologues in Ciona intestinalis, respectively. Moreover, one other IRK gene existed in the tunicates. Second, the three types of tunicate IRK, elementary IRK, GIRK, and possibly ATP-regulated IRK, were representatives of three major groups of the human IRK proteins. All protostomic IRKs were grouped differently from Halocynthia IRKs, though they may show some similarity to a putative Ciona ATP-regulated type IRK. Prokaryotic IRKs are all evolutionally different from those eukaryotic ones. To illustrate the major phylogenetic evolution of IRK genes more simply and clearly, we aligned the AA sequences of tunicate, Caenorhabditis, and bacterial IRK proteins, and the tree was made, as shown in Fig. 4. All internal branches in the tree were verified by a Bootstrap test, suggesting that the elementary IRK group and G-protein-activated group diversified from the putative ATP-regulated type in the tunicate clade later than the deuterostomic and protostomic diversification in the ancestor IRKs, possibly one of the putative ATP-regulated types. The results further confirmed that the IRKs of protostomic clade were different from those of the deuterostomic clade, but they evolved after the eukaryotic diversification from prokaryotic evolution.
Comparison of intron exon structures of IRK genes between protostomic and deuterostomic clades
As described above, 6 to 11 introns were inserted in the coding regions of the Halocynthia IRK genes. However, 108 IRK genes of the vertebrate clade, which are extracted from recently established genomic databases listed in Additional file 5 [see Additional file 5], revealed relatively few or no introns in their coding regions. Figure 5 illustrates frequency histograms of intron numbers in all IRK genes examined. Figure 5A, 5B, and 5C represents the histograms for the three major groups of IRKs of the vertebrates, that is, ATP-regulated, elementary, and G-protein-activated IRKs, respectively. The results indicate that the intron numbers were significantly larger in the tunicate IRKs, but they were rather few either in the amniotic or anamniotic vertebrates, though the anamniotic vertebrates might show some larger numbers, especially in the case of ATP-regulated IRKs. For the protostomic clade, the Caenorhabditis definitely revealed the large number of 9, but the Drosophila and Anopheles revealed relatively small numbers of less than 4, as shown in Fig. 5D. In Figure 6 and 7, 7 tunicate, 6 Caenorhabditis, and 7 bacterial IRK AA sequences were aligned, and the portions of intron insertion are illustrated [see Additional file 6]. In either tunicates or Caenorhabditis, the sites for insertions were strictly conserved between the orthologous protein pairs, at least in the major feature regions of IRK genes. As in the case of tunicates, the Caenorhabditis IRK genes revealed insertion sites at the physiologically important membrane regions and the ion pore regions. The further three common insertion sites were found between the tunicates and the Caenorhabditis (Fig. 6 and 7, yellow background and boxed characters). Thus in terms of the AA sequences, the Caenorhabditis and the tunicate IRKs were separately grouped and evolved differently, but the macroscopic genomic structures seemed to keep commonality between them.
Naturally bacterial IRK protein AA sequences show no intron insertion, but in alignment with eukaryotic IRK AA sequences, they revealed six intrinsic gaps in their conserved and feature regions as the IRK sequences, showing that no eukaryotic equivalent sequences existed in the modern prokaryotes. However, it should be noted that either edge of the respective gap seemed to correspond at least to some of the intron insertion sites in the Caenorhabditis genomes in the five cases out of these six gaps (Fig. 6 and 7; purple background characters in the bacterial AA sequences).
Comparison of AA sequences and genomic structures among 7 tunicate IRK genes and 108 vertebrate IRK genes
In Figure 8, the phylogenetic tree obtained from the alignment of a total of 115 IRK AA sequences of tunicates and vertebrates is illustrated. As shown above in the tree composed of human and tunicate sequences, the 3 major functional groups of IRK proteins are exactly represented in the major branches of the tree composed from 115 IRK sequences. And it was concluded that the 2 representative IRKs of tunicate GIRKs, TuGIRKAa and B, and the elementary IRK, TuIRKA, must be descendants of the ancestor forms of vertebrate GIRK and IRK, respectively. Although grouping the putative Ciona ATP-regulated IRK as a descendant of the ancestor sequence for ATP-regulated IRKs, including ROMKs, was not supported by a Bootstrap test, there were suggestive data to assign the common ancestor for all vertebrate IRKs to one of the relatives of the ATP-regulated types. From the survey of the genomic databases, a novel evolutional correlation was found between ATP-regulated IRK genes and G-protein-activated IRK genes, revealing the antitandem or palindromic duplication (Table 1).
On the tree derived from the AA sequences, the GIRK subbranch was distantly located from the subbranch of ATP-regulated IRKs. However, as illustrated in Table 1, the genome structures of respective IRKs in the one subbranch had their partners in the other subbranch, and each pair was located very closely together on the same chromosome in an antitandem or a palindromic way. For example, the 5' terminals of Kcnj 1 or ROMK1 genomes of human, mouse, rat, and chick were only 20 to 80 Kbp apart from the 5' terminals of Kcnj5 or GIRK2 genomes of respective species, and their 5' to 3' directions were just opposite to each other (green characters enclosed by interrupted lines in Fig. 8). There were two other similarly paired subbranches of Kcnj10 and Kcnj9 (orange characters) and of Kcnj15 and Kcnj6 (dark blue characters). The separation distances on the respective chromosomes were 10 to 20 Kbp for the former pairs and 260 to 450 Kbp for the latter. This type of pairing could be expected to apply also to the remaining subbranches of Kcnj 3 and Kcnj13. The genomes of the Kcnj 3 subbranch were all located on the equivalent chromosomes, where HoxDs were located in human, mouse, rat, and chick genomes, but only human Kcnj13 gene was located on the same chromosome as that of HoxD, with a relatively large separation distance from the human Kcnj3 genome.
These results suggested that both ancestors of GIRK and ATP-regulated IRK genes were antitandemly or palindromically aligned duplicates of each other possibly on the chromosome of ancestor chordates. After the event, two round duplications of chromosome numbers in the ancestor vertebrates  made approximately 4 times the members of genes in each major branch in the vertebrate clade, as suggested in the case of HOX genes . As also suggested in the case of HOX genes, there may be some extra chromosome duplications in teleost fish or amphibians , because we could find some extra members of the IRK genes of puffer fish, zebra fish, and Xenopus in the respective subbranches. A recently reported Ciona genomic database from Ensembl suggested that the Ciona ATP-regulated IRK gene (Protein, ENSCINP00000010507; gene, ENSCING00000005094) is located at Chromosome 4q-373, but there was no information yet on the location of the Ciona G-protein-activated IRK gene (JGI; ci0100138554). The functional significance for the preservation of the proximity of paired IRK genomes on the vertebrate chromosomes will be discussed later.
There was also evidence of genuine tandem duplication within the branch of the elementary IRK gene group, as shown in the paired Kcnj2 and Kcnj16 subbranches (Fig. 8; Table 1). In this case, the directions from the 5' terminal to the 3' terminal were the same within respective pairs. This tandem duplication in the mouse IRK genome was noted previously , and this pairing was considered to be of an origin different from the above antitandem pairings. No other evidence of pairing was found in the elementary IRK gene group, except that a possible weak linkage between Kcnj12 and Kcnj2 subbranches was found as listed chromosomal location in Additional file 5 [see Additional file 5].
A uniquely conserved intron insertion site among the vertebrate GIRK gene group and tunicate GIRK genes
As described above, intron insertions in the vertebrate IRK coding regions are very few in comparison with those of the tunicate IRKs. Thus the intron and exon structures of the vertebrate IRK genes were quite different from those of the tunicates, though the AA sequences are phylogenetically closely related to their tunicate orthologs, as shown in the tree in Fig. 8. However, in the group of GIRK genes there is only one uniquely conserved intron insertion site among all vertebrate GIRKs and tunicate GIRKs in Fig. 9 and Fig. 10 [see Additional file 7]. As shown Fig. 11, this site was between the first and second nucleotide in the glycine codon and was preserved exclusively in the group of vertebrate and tunicate GIRKs. The site corresponded to G307 residue in the mouse GIRK1 or Kcnj3 and was located at the loop region near the hydrogen bonded turn a304–306 and between two beta strands a294–303 and a309. The site was suggested to be in the hinge region in the cytoplasmic C-terminal domain, which could be important for open-close transition by G-protein beta and gamma binding, and the residue seemed to be exposed to the cytoplasmic surface, as inferred from the mouse GIRK1 cytoplasmic crystal structure reported by Nishida and MacKinnon. It is also noted that two exclusively conserved AA sequence regions in the C-terminal domain of the GIRK were located in the upper stream and downstream within 75 and 25 AA residue distances, respectively, from this intron insertion site and. The former, a234–237, were included in the beta-strand region a223–237 and seemed to be located on the cytoplasmic surface(Fig. 12). The latter, L333, was previously suggested to be the most effective activation site by G-protein beta and gamma (Fig. 13) . The above-described conserved three regions seemed to be all located on the probable G-protein binding surface, making the solid triangle of molecular flame for the binding surface. The conserved intron insertion site, G307, seemed to be an especially important hinge region by which the working angle could be used to adapt to the global configuration of the binding surface.
Phylogenic analysis and comparison of genomic structures of G-protein beta(GNB)s
Since the intron insertion site, which was supposed to be in the G-protein beta binding region, was exclusively conserved in the group of GIRK genes, we attempted to analyze the genomic structures of G-protein betas interacting with these GIRKs. The data were all obtained from the established genomic databases [see Additional file 8]. Figure 14 illustrates the phylogenetic tree derived from 54 G-protein beta AA sequences and 7 bacterial WD-repeat protein AA sequences as the outgroup. Because this case was different from the IRK sequences, two major groups of G-protein beta very likely existed within the ancestor sequences of both protostomic and deuterostomic clades, that is, the first group of GNB1 to 4 and the second group of GNB5. In the vertebrate clade after diversification from the ancestor chordate, the first group produced 4 members, and the number seemed to remain throughout the vertebrate evolution, except for some extra production in the teleost fish. Also in fly and mosquito, two major groups remained.
In terms of intron-exon structures, the frequency histograms in Fig. 15 demonstrated that the 7 to 11 intron insertions in the coding region were rather constant among all vertebrate GNBs and Ciona GNBs. Two groups of GNBs in Caenorhabditis revealed 8 and 7 intron insertions, similar to those in Ciona, respectively. However, the GNBs in fly and mosquito revealed a much lower number of intron insertions than Caenorhabditis did. An alignment of GNB1 protein sequences illustrated not only strict amino acid residue conservation, but also strict intron insertion site preservation among human, mouse, rat, fish, Ciona, and Caenorhabditis GNB1 proteins (Fig. 16 and 17). However, fly and mosquito GNBs did not conserve the intron insertion sites. The alignment of GNB5 protein AA sequences illustrated similar results [see Additional file 9]. The physiological significance of intron insertion site conservation will be discussed later.
Summary of results
We sequenced three Halocynthia IRK genomic genes and determined their genomic structures. Furthermore, a total of 131 IRK genomic genes were inferred from the established genomic databases. The phylogenic trees derived from the known or inferred AA sequences revealed the clear diversification of deuterostomic IRK from protostomic IRK, and the tunicate IRKs were possibly representatives of ancestor forms of three major groups of IRKs in the vertebrate. However, the genomic structures, including intron-exon connections of the tunicate IRKs, showed considerable similarities to those of Caenorhabditis. Although the major groups of tunicate IRK proteins included only one or two members, in the vertebrate clade the members in one major group increased at least four times. The generation of paralogues can be achieved by various types of gene duplication, such as tandem and/or chromosomal. However, the generation of the major groups in the tunicates and vertebrates may be of more ancient origin, and some of the mechanism may be derived from the unique palindromic duplication. Finally the strong conservation of intron insertion sites in the coding regions of vertebrate GIRK group indicated the close functional correlation between the genome structures and the protein structures.
Intron early theory
The significance of intron insertion has been considerably notified by the claim of exon theory . The exon theory persisted that the ancestral eukaryotic genes are generated from numerous short exons connected with ancient introns, and exon shuffling introduced new functions into the gene product proteins . Provocative evidence for the theory is provided by the high-frequency insertion of introns between codon sequences, phase 0 intron, and the conservation of intron insertion sites in the coding regions of orthologous proteins from a phylogenetically wide range of taxis . However, because of massive intron loss during the evolutionary process and the significant production of new intron insertion in the highly evolved taxis, such as human and Drosophila, the ancient intron insertion sites may be vastly lost in the genomes of present-day organisms . Those situations have facilitated several types of late intron theory or synthetic theory of intron evolution [10–12].
In the present study, we purposefully limited observation and discussion to within the gene group of definitely functionally defined proteins, that is, IRK channels. Results indicated that both the tunicate IRK genes and the Caenorhabditis IRK genes conserved large numbers of intron insertion sites in the functionally important transmembrane and pore regions. Further, three insertion sites were commonly conserved in terms of AA sequences between the tunicates and the Caenorhabditis, indicating that some of the cornerstones of early evolutional history remained.
Although there is an intriguing hypothesis that the exon corresponds to protein domains [23, 36], the conserved insertion sites seemed not always to exist at the borders of domains. Instead, they were at the structurally critical points, such as the membrane spanning region or inside the pore regions in the present study. This point will be further discussed below.
In both chordates and protostomal clades, when the respective members increased within functionally different major groups of IRK genes, the number of intron insertions abruptly decreased. In the tunicates or the Caenorhabditis, the members remained small in number possibly because of the size constraint of whole genomes, and the intron insertion sites were conserved. Our observation could suggest no points about intron insertions in bacterial genes; however, we were inclined to consider that the intron appearance is closely related to the early appearance of eukaryotic organisms, because some residual gaps of IRK amino acid sequences were left in the modern bacteria, roughly corresponding to the conserved intron insertion sites of the ancestral eukaryotes reflected in the Caenorhabditis or tunicate genomes. This suggested that the intron insertion mechanism newly appeared at least at the time of diversification between the eukaryotic and prokaryotic ancestors, or that the introns in the prokaryotic and eukaryotic common ancestor, the progenote, genomes disappeared in the prokaryotes, as suggested previously . This consideration supported the exon theory in the evolution of the IRK gene family.
There remains the possibility that the common intron insertion between tunicates and Caenorhabditis could be derived from convergent evolution during later history way after eukaryotic diversification from prokaryotes. However, it was totally unclear what kinds of adaptive merits produced by the multiple intron insertion commonly existed in their phylogenetically different ancestors, although convergent evolution of the AA sequences might be derived from the functional similarity of IRK proteins .
In the vertebrate it has been proposed that gene diversification has been attained by various kinds of gene duplication [15, 16, 19, 38]. There are many examples of tandem duplication, which could produce the paralogues of genes . It could be inferred that one of the duplicated pair became free from the functional constraining and acquired the new function or degenerated as nonfunctional by the accumulation of mutations. It is generally understood that the duplication by random insertion could be derived from the genomic insertion of reverse transcript of the original gene by various transposon-like mechanisms and that the exact duplication of gene regulatory sequence, such as promoters and enhancers, are rather difficult. Thus block duplication or chromosomal duplication of the concerned gene groups should be an easier way to produce new functional paralogues. The famous two-round duplication of the chromosome set has been proposed by Ohno in the ancestor vertebrate , and the theory has strong support by the existence of four sets of tandemly aligned Hox gene groups in the amniotic vertebrate genomes and the recent analysis on the whole genome sequences derived from a wide-range of vertebrate taxa including both anamniotes and amniotes [14–16]. The presence of extra sets of Hox gene groups in fish has been explained by the extra duplication of chromosomal sets in the teleosts.
In the present study, two major gene groups, ATP-regulated IRK and G-protein-activated IRK, of which gene branches were placed quite distantly on the phylogenetic tree derived from the IRK AA sequences, were found to be physically correlated closely and franked anti-tandemly as pairs on the individual chromosomes. This is so-called palindromic duplication, which has recently attracted much attention in the adaptive gene expansion of carcinoma stem cell or fungi, Saccharomyces, genomes [18–20, 38]. The unique palindromic duplication during evolutional time found in the present study would forcibly make the gene pair share the gene regulatory regions. Although the promoters may be separately placed, the coincidental transcriptional activity should be difficult between the gene pair. The expression must be mutually exclusive. Actually, G-protein-activated IRKs are exclusively expressed in the neuronal cells of the nervous tissues and heart cells , and the paired ATP-regulated IRKs are expressed in the epithelial cells and the glia of the nervous tissue [41–45].
This rather unique type of palindromic duplication must have been originated before the subbranching in respective major groups of genes, that is, at the time of the ancestors common to the tunicate and the vertebrates. Then the parallel but independent functional evolution during long archeological history could occur in the gene pairs deriving, respectively, two major distinct gene groups as a result of mutual and physical constraining in the gene regulatory regions. Therefore the palindromic duplication was highly expected to exist also in the modern tunicate genome. However, no evidence of this type duplication of IRK genes has remained in the genome though the physical maps of the tunicate chromosomes have not yet been completed as described in the Results section. In the Ascidia, the tandem arrangement of Hox genes was dispersed, as previously reported [46, 47]. Similarly, it was possible that the evolutional evidence of the antitandem or palindromic duplication of the IRK genes may disappear because of the gene dispersion in the Ascidia during the evolutionary time. As Ferrier and Holland  have suggested, the dispersion of the Hox gene groups may be deduced from the inference that the temporal colinearity as well as constraint on the cluster organization might have been removed from animals that undergo rapid embryogenesis with a low cell number and predominantly mosaic development. However, it may also be plausible that the dispersion in the genome itself could lead the rapid embryogenesis with a low cell number and predominantly mosaic development, resulting from the removal of the spatial and temporal colinearity in the genomic structure and regulation, without which the animals had to become adaptable.
Two rather marginal but important points could be discussed in relation to the palindromic duplication. First, according to the constructed phylogenetic tree the G-protein-activated IRK gene may be diverged from the ATP-regulated IRK gene. After the palindromic duplication of ancestoral ATP-regulated IRK genes, the G-protein binding segment on the duplicated genes could be forced to evolve in parallel with G-protein beta and gamma genes. On the other hand, the coupled evolution between G-protein-activated and ATP-regulated IRKs as a result of the possible joint ownership of the gene regulatory regions must be related to apparently independent but actually interdependent functions of the epithelial system derived from mesenchymal or endodermal sheet and the tubular nervous system from ectodermal sheet in the common ancestor of the tunicates and vertebrates. Ionic balance, especially potassium, resulting from the IRK function in the epithelial transport systems and neural excitable epithelium must have been important adaptive factors to coordinate the regulations of K ion transport activities of kidney and neurons in the freshwater or landed vertebrate evolution to balance their effects on the internal milieu and membrane excitability. For example, the high activity of ROMK in the kidney, one of the members of ATP-regulated IRK, will produce low K+ concentration in the internal milieu, and, if occurring simultaneously, the high activity of GIRKs in neurons or heart cells will be extremely inhibitory for membrane excitability. In the above respect, it was interesting that the seawater living tunicates have a dispersed location between the ATP-regulated and G-protein activated IRK genes, where the ionic internal milieu is balanced mainly by the seawater composition. In this discussion, it was important to notice that the ATP-sensitive or SUR-coupled IRKs, such as Kcnj8 and Kcnj11, were evolutionally different from ATP-regulated IRKs, such as ROMK (see Fig. 5), and that ATP-sensitive IRK genes were not paired with G-protein-activated IRK genes. Actually ATP-sensitive IRK did coexist with GIRKs in the heart cells and neurons.
Second, by closely looking at the phylogenetic tree, we see that the branching in the respective subbranches was not exactly parallel between the two paired major IRK gene branches, as shown in Fig. 8. The exchange of partners resulting from one round of homologous recombination between the two members of gene pairs within four members during the two-round chromosomal duplications nicely explained the slight difference of the subbranching pattern between two major branches of ATP-regulated and G-protein-activated IRK groups.
Functional significance of intron insertion
Most introns in the IRK genomes disappeared in the vertebrate genomes after gene diversification. However, unique intron insertion sites remained precisely conserved in respect to both AA code and intracodon insertion sites exclusively in the G-protein-activated IRK gene group, including the tunicate ones. As AA sequences, the conservation of the site was partially conceived in the genes of other major groups. The site is the hinge region that is important for the opening-closing of the IRK channel correlated with G-protein beta and gamma binding, as described in the Results. The exclusively conserved Glycine, which was also the conserved intron insertion site, composed the hinge region that must bend or rotate as the pivotal role in adjusting against the G-protein beta and gamma binding and accordingly to change the open or closed state of channels. Thus the strongest evolutionary conservative pressure must be applied to the hinge site. Here it could be inferred that the conservation of intron insertion with phase 1 but not phase 0 or 2, which has a redundancy at the 5' or 3' side of introns, was probably the most plausible mechanism to result in an all-or-none or alternative mutation, that is, conservation or fetal. This is because any addition, deletion, or point mutation in the neighborhood of the insertion site would produce mostly a disruption of intron insertions, or even in rare cases a dysfunctional change in an amino acid residue, even if the intron insertions were kept. Thus it could be easily imagined that both sides of the conserved phase 1 intron insertion site could be confronted by the high evolutional conservative pressure or strict parallel evolution to be adaptable for binding to the partner proteins.
In the above, it was interesting that our survey on the genome structures of G-protein betas (GNBs), of which GNB1 is especially the well-known partner of GIRKs and also other many G-protein activated proteins [48–50], revealed the extremely conserved intron insertion sites from the tunicate genes to the human genes in the vertebrate clade. Among the GNB genes, the vertebrate evolution seemed to introduce the functional diversification of only marginal activity by the member increase because of whole genome duplication in the specialized groups on the phylogenic tree. It was suggested instead that because of the multi-interactive nature of the protein, the high conservative pressure on the pivotal sites in their molecular structure made their intron insertion sites, including AA codes, constant during the evolutional time. Actually, the conserved sites of phase 1 intron insertion are placed on the hinge regions between two beta sheets or helix structures, as inferred from rat GNB1 crystal structures reported [51, 52]. Thus it could be inferred that the site conservation of intron insertion, especially phase 1 intron, is one of the representatives of the genome structures constrained by the protein structures for the adaptation to the protein-protein interaction. On the other hand, it has been well discussed that phase 0 or phase 2 introns have a significant role in exon shuffling or domain exchange for protein structural evolution .
Structure and evolutionary rate
It has been hotly debated that protein structural constraints because of protein-protein interaction could introduce a slow evolutionary rate in AA sequences of the contacting regions of both proteins, because of the necessity of the preservation of functionally significant structures and of the parallel evolution between the paired interacting regions [53–55]. This inference was also reasonably conceived in the C-terminal regions of the IRKs in comparison with the N-terminal and membrane region of the IRKs. As shown in Fig. 18 and 19, two separate phylogenetic trees for the IRK N terminal and membrane-spanning region and for the IRK C terminal region illustrate that the C-terminal region of G-protein activated IRK group revealed rather slower evolutionary rates or lower mutational rates than those of ATP-regulated IRK group except the Kcnj1 subgroup. The N-terminal and membrane-spanning regions of the ATP-regulated IRKs revealed slower evolutionary rates than those of the GIRK. This may be derived from the protein-protein or protein-ligand interactions occurring in the intramolecularly different regions between the GIRKs and the ATP-regulated IRKs. The exceptional Kcnj1 subgroup, which revealed a slower rate in the C-terminal region, has been known to have the ATP/PIP2-binding site in a 39-amino-acid region of the C-terminal . Thus it was suggested that the structural constraints of the proteins limit the evolutionary rates of the particular protein regions. The conservation of intron-insertion sites at the functionally significant regions described above could be one of the important mechanisms to facilitate this general tendency.
In the present study, we attempted to analyze the genomic structures of a group of proteins, which are functionally well defined, and to compare them among a phylogenetically wide range from protostomal to deuterostomal clades. We aimed to elucidate the functional significance in evolutional changes in the genome structure, such as intron insertion and gene duplication, and the interactive relation between the expressed protein structure and the genomic structural changes. The results follow. First, the general inron-exon structures may be derived from the early time of eukaryotic evolution, when the protein family revealed the first and elemental function with their unique molecular structures. Second, the various types of gene duplication could contribute to the gene diversification resulting in the functional evolution, but some constraints in the genomic structures because the unique molecular mechanisms of the duplication, such as palindromic duplication, may produce the adaptive parallel or coordinate evolution among the protein family members between organs or tissues that were functioning differently. Third, the conserved intron insertion site, especially in the case of phase 1 intron, could be correlated with amino acid residues that were structurally pivotal roles in the protein-protein interaction. Thus it was suggested that the ancient intron insertion sites, which are intrinsically related to the functionally key region of the proteins, in the eukaryotic genome, or possibly in the prokaryotic and eukaryotic common ancestor, and the unique palindromic genomic duplication in the genomes of vertebrate clade shaped an elementally functioning membrane protein family during evolution.
Halocynthia roretzi (HR) IRK cDNA
The cDNAs, TuIRKA, TuGIRKAa, and TuGIRKB were gifts from Dr. Y. Kubo in the Division of Biophysics and Neurobiology, Department of Molecular Physiology, National Institute for Physiological Sciences, and were subcloned into pBluescript II SK2 (Stratagene, San Diego, CA) in accordance with the manufacturer's protocol and sequenced [see Additional file 2].
Genomic library screening
A Halocynthia roretzi genomic library constructed in λEMBL3/BamH I Vector (Stratagene) was a gift from Dr. H. Okado of the Department of Molecular Physiology, Function Research Division, Tokyo Metropolitan Institute for Neurosciences. This library was screened using specific EcoRI fragments of respective HR IRK cDNAs as probes. Filters were prehybridized in a solution containing 50% formamide, 53 SSC, 53 Denhardt's solution, and 0.1% SDS for 1 h at 42°C, then hybridized overnight at 42°C in the same solution containing Digoxigenin-labeled probes. The DNA probes were random primed, labeled with Digoxigenin-11-dUTP using DIG-High Prime, a 5 × concentrated labeling mixture of random hexamers, dNTP mix containing alkali-labile Digoxigenin-11-dUTP, and labeling grade Klenow enzyme, and optimized according to the manufacturer's manual (Roche; DIG-High Prime DNA Labeling and Detection Starter Kit II). The filters were washed twice in 23 SSC and 0.05% SDS at room temperature for 20 min and once in 13 SSC and 0.1% SDS at 55°C for 2 h. They were then treated with chemiluminescence reaction buffer (Roche; DIG-High Prime DNA Labeling and Detection Starter Kit II). Positive clones were detected by exposing the luminescence to X-ray films contacted with the filters. The λEmble3 clones containing these positive inserts were amplified and used for the genomic sequencing. The obtained clones were 2 for the TuIRKA gene, the TuIRKA11 clone, and the TuIRKA14 clone, 2 for the TuGIRKAa gene, the TuGIRKAa3 clone, and the TuGIRKAa9 clone, and 1 for the TuGIRKB gene and the TuGIRKB2 clone, showing the possible coverage of all ORFs, 5' UTRs, and 3' UTRs with restriction maps [see Additional file 3]. And later the coverage was all confirmed with the sequencing. These 5 clones corresponding to respective cDNAs were digested with Xho I and Sal I, and the digests were subcloned into pBluescript II SK2 using an Xho I or Sal I restriction site, respectively. For the TuIRKA gene, the obtained subclones were 17 from the first λEMBLE3 clone and the TuIRKA11 clone and 3 from the second λEMBLE3 clone and the TuIRKA14 clone. For the TuGIRKAa gene, 7 subclones from the TuGIRKAa3 clone and 3 from the TuGIRKAa9 clone were obtained. For TuGIRKB, the subclone was recloned into pBluescript II SK2 from the TuGIRKB2 clone itself. All subclones were sequenced with appropriate primer combinations at least in both directions, and in some cases a third sequencing was carried out. In the case of the TuGIRKA gene, PCR clonings of the two gap regions between subclones were performed on the original λEMBLE3 clones, using part of the decided flanking sequences as primers, and also subcloned into pBluescript II SK2 and sequenced. All suspected boundaries between the neighboring subclones were subjected to perform PCR on the original λEMBLE3 clones, using appropriate primers for the tagged sequence derived from the sequenced regions placed at both ends of the boundaries. And the continuation beyond the boundaries of subcloned fragments were all confirmed.
DNA was sequenced by the Dye Terminator method, using an automated laser fluorescent DNA sequencer (ABI PRISM model 3100, Applied Biosystems, Foster city CA, USA) and a BigDye Terminator Cycle Sequencing Kit (Applied Biosystems, Foster city CA, USA). DNA sequences were analyzed by using the computer program DNASIS3.2-MAC (Hitachi Software, Tokyo, Japan). Exon-intron boundaries were identified by comparing the genomic and cDNA sequences. All obtained genomic and cDNA sequences necessary for the present experimental conclusions are listed in Additional files 1, 2, 3 and 4 [see Additional files 1, 2, 3 and 4].
Retrieving sequences from the C. intestinalis genome and a cDNA/EST database
The C. intestinalis IRK channel protein sequences were tBLASTn searched against the draft or completed genome sequence (Ref.: , and Ensembl and DOE Joint Genome Institute [JGI] Ciona intestinalis v2.0 [JGI site for the complete C. intestinalis genome sequence and gene annotations]: ) and a cDNA/EST database (Ref.: , and Ghost Database: ) using human and HR IRK channel protein AA sequences [29, 30]. IRK channel proteins were identified by using the following basic method. Briefly, when the corresponding cDNA sequence covering the diagnostic sequences for an IRK channel molecule, such as a channel pore region and two franking transmembrane regions, and available by InterPro search , the deduced protein sequence was used for the analyses. When the cDNA sequence was not available and grail EXP or genewise confidently predicted the gene encompassing the entire channel region, the peptide sequence deduced from the gene model was used. When the predicted gene model was not perfect, but the ESTs covered either the entire region or the region lacking the gene model, the peptide sequence was deduced from the assembled sequence obtained by using either a set of ESTs (5' and 3' EST pair), multiple sets of ESTs, or both an EST and the gene model. After the protein AA sequence was identified, the respective genomic sequences were retrieved from databases. All analyzed Ciona IRK genomic genes and other genes are listed in Additional file 5 [see Additional file 5].
Molecular phylogenetic analysis
The identified or predicted IRK channel protein sequences from human (database version: NCBI Build 36.2), mouse (database version: NCBI Build 36.1), Takifugu rubripes (database version: IMCB/JGI FUGU 4.0), zebra fish (Danio rerio, database version: Ensembl Zv 6), rat (Rattus norvegicus, database version: RGSC v3.4), chicken (Gallus gallus, database version: WASHUC 1), and frog (Xenopus tropicalis, database version: JGI 4.1) were retrieved from NCBI Genomes  and/or Ensembl . Diagnostic sequences for each retrieved IRK channel molecule, such as a channel pore region and two franking transmembrane regions, were identified by InterPro search . To delineate vertebrate gene families, a similarity search was performed (tBLASTn, ; E-value cutoff E-10) with all IRK channel proteins from the organisms listed in Additional file 5 [see Additional file 5], that is, the proteins of HR and Ciona intestinalis (database version JGI2.0), Drosophila melanogaster (database version BDGP 4), Caenorhabditis elegans (database version WS 150), Anopheles gambiae (database version AgamP3), and seven prokaryotic IRK channel proteins that were added as the outgroup species (NCBI Genomes Prokaryotic Projects). The prokaryotic sequence was used to root the phylogenetic tree. Redundancy between the families was removed. The sequences were aligned by using the ClustalX 1.83 program [63, 64]. Alignments in nonconserved regions of channel proteins were carefully checked by eye, and regions with ambiguous alignments, in particular at the less-conserved cytoplasmic region and extracellular region, were eliminated. Thus verified alignments were used to construct phylogenetic trees. Neighboring joining trees (with 500 bootstrap replicates) were constructed using MEGA3 [65, 66] by using a substitution model of Amino: Poisson correction with Gamma Distributed Rates among sites (Gamma Parameter 2.0) and including sites with a pairwise deletion of Gaps/Missing data. For the analyses shown in Figs. 3, 4, 8, 18 and 19 when the NJ trees were constructed, branch points were considered significant only when the bootstrap tests gave a significant value >50%. The sequences used are designated in succession by the abbreviation of the species and the gene name. Abbreviations of the species are as follows: HR for Halocynthia roretzi, CI for Ciona intestinalis, HS for human, MM for M. musculus, RN for Rattus norvegicus, GG for Gallus gallus, DR for Danio rerio, FR for Takifugu rubripes, XT for Xenopus tropicalis, DM for D. melanogaster, AG for Anopheles gambiae, CE for C. elegans, and CB for C. briggsae. After the protein AA sequences were identified and the trees obtained, all respective genomic sequences were retrieved from the databases. And the inferred sites of exon intron junctions and the length of respective introns were used for a comparison of genomic structures among the organisms examined.
Relative dating of duplication events
The presence of various palindromic gene duplication events was systematically analyzed on the Phylogenetic trees with standard distance metrics obtained above. Duplication events were evaluated by relative dating and thus were based on the relative position of the duplicated genes compared to speciation events in the phylogenetic tree.
Retrieving G-protein beta (GNB) AA sequences and their genome structures from the genome databases
A total of 54 AA sequences of eukaryotic G-protein beta and two beta-like AA sequences were inferred from genome databases, NCBI genome projects, and Ensembl. To delineate GNB gene families, a similarity search was performed by using human G-protein beta1 and beta5 AA sequences (tBLASTn, ; E-value cutoff E-10), and all GNB AA sequences retrieved from both protostomal and deuterostomal clades were listed in Additional file 8 [see Additional file 8]. Moreover, seven bacterial WD-repeat protein AA sequences were retrieved from the NCBI prokaryotic genome data projects and were used as the outgroup for the construction of GNB trees. After the G-protein beta AA sequences were identified and the trees obtained, all respective genomic sequences were also retrieved from the databases. And the exon intron junctions, length of introns, and numbers of intron insertions within coding regions were inferred, as listed in Additional file 8 [see Additional file 8]. Tunicate GNB data were obtained from the JGI database, Ciona intestinalis v2.0. Alignment was carried out by the ClustalX 1.83 program [see Additional file 9]. The phylogenetic tree derived from the aligned data was made by the Neighbor Joining Method in the Mega3 v3.1 program, using the Amino Poisson correction model with Gamma-distributed rates among sites.
Benson DA, Karsch-Mizrachi I, Lipman DJ, Ostell J, Wheeler DL: GenBank. Nucleic acids research. 2006, 34 (Database issue): D16-20. 10.1093/nar/gkj157.
Birney E, Andrews D, Caccamo M, Chen Y, Clarke L, Coates G, Cox T, Cunningham F, Curwen V, Cutts T, Down T, Durbin R, Fernandez-Suarez XM, Flicek P, Graf S, Hammond M, Herrero J, Howe K, Iyer V, Jekosch K, Kahari A, Kasprzyk A, Keefe D, Kokocinski F, Kulesha E, London D, Longden I, Melsopp C, Meidl P, Overduin B, Parker A, Proctor G, Prlic A, Rae M, Rios D, Redmond S, Schuster M, Sealy I, Searle S, Severin J, Slater G, Smedley D, Smith J, Stabenau A, Stalker J, Trevanion S, Ureta-Vidal A, Vogel J, White S, Woodwark C, Hubbard TJ: Ensembl 2006. Nucleic acids research. 2006, 34 (Database issue): D556-61. 10.1093/nar/gkj133.
Okubo K, Sugawara H, Gojobori T, Tateno Y: DDBJ in preparation for overview of research activities behind data submissions. Nucleic acids research. 2006, 34 (Database issue): D6-9. 10.1093/nar/gkj111.
Dehal P, Satou Y, Campbell RK, Chapman J, Degnan B, De Tomaso A, Davidson B, Di Gregorio A, Gelpke M, Goodstein DM, Harafuji N, Hastings KE, Ho I, Hotta K, Huang W, Kawashima T, Lemaire P, Martinez D, Meinertzhagen IA, Necula S, Nonaka M, Putnam N, Rash S, Saiga H, Satake M, Terry A, Yamada L, Wang HG, Awazu S, Azumi K, Boore J, Branno M, Chin-Bow S, DeSantis R, Doyle S, Francino P, Keys DN, Haga S, Hayashi H, Hino K, Imai KS, Inaba K, Kano S, Kobayashi K, Kobayashi M, Lee BI, Makabe KW, Manohar C, Matassi G, Medina M, Mochizuki Y, Mount S, Morishita T, Miura S, Nakayama A, Nishizaka S, Nomoto H, Ohta F, Oishi K, Rigoutsos I, Sano M, Sasaki A, Sasakura Y, Shoguchi E, Shin-i T, Spagnuolo A, Stainier D, Suzuki MM, Tassy O, Takatori N, Tokuoka M, Yagi K, Yoshizaki F, Wada S, Zhang C, Hyatt PD, Larimer F, Detter C, Doggett N, Glavina T, Hawkins T, Richardson P, Lucas S, Kohara Y, Levine M, Satoh N, Rokhsar DS: The draft genome of Ciona intestinalis: insights into chordate and vertebrate origins. Science (New York, NY. 2002, 298 (5601): 2157-2167.
Cochrane G, Aldebert P, Althorpe N, Andersson M, Baker W, Baldwin A, Bates K, Bhattacharyya S, Browne P, van den Broek A, Castro M, Duggan K, Eberhardt R, Faruque N, Gamble J, Kanz C, Kulikova T, Lee C, Leinonen R, Lin Q, Lombard V, Lopez R, McHale M, McWilliam H, Mukherjee G, Nardone F, Pastor MP, Sobhany S, Stoehr P, Tzouvara K, Vaughan R, Wu D, Zhu W, Apweiler R: EMBL Nucleotide Sequence Database: developments in 2005. Nucleic acids research. 2006, 34 (Database issue): D10-5. 10.1093/nar/gkj130.
Wheeler DL, Barrett T, Benson DA, Bryant SH, Canese K, Chetvernin V, Church DM, DiCuccio M, Edgar R, Federhen S, Geer LY, Helmberg W, Kapustin Y, Kenton DL, Khovayko O, Lipman DJ, Madden TL, Maglott DR, Ostell J, Pruitt KD, Schuler GD, Schriml LM, Sequeira E, Sherry ST, Sirotkin K, Souvorov A, Starchenko G, Suzek TO, Tatusov R, Tatusova TA, Wagner L, Yaschenko E: Database resources of the National Center for Biotechnology Information. Nucleic acids research. 2006, 34 (Database issue): D173-80. 10.1093/nar/gkj158.
Satou Y, Takatori N, Fujiwara S, Nishikata T, Saiga H, Kusakabe T, Shin-i T, Kohara Y, Satoh N: Ciona intestinalis cDNA projects: expressed sequence tag analyses and gene expression profiles during embryogenesis. Gene. 2002, 287 (1-2): 83-96.
Roy SW, Gilbert W: Complex early genes. Proceedings of the National Academy of Sciences of the United States of America. 2005, 102 (6): 1986-1991. 10.1073/pnas.0408355101.
Roy SW, Gilbert W: The pattern of intron loss. Proceedings of the National Academy of Sciences of the United States of America. 2005, 102 (3): 713-718. 10.1073/pnas.0408274102.
Rzhetsky A, Ayala FJ, Hsu LC, Chang C, Yoshida A: Exon/intron structure of aldehyde dehydrogenase genes supports the "introns-late" theory. Proceedings of the National Academy of Sciences of the United States of America. 1997, 94 (13): 6820-6825. 10.1073/pnas.94.13.6820.
Venkatesh B, Ning Y, Brenner S: Late changes in spliceosomal introns define clades in vertebrate evolution. Proceedings of the National Academy of Sciences of the United States of America. 1999, 96 (18): 10267-10271. 10.1073/pnas.96.18.10267.
Yoshihama M, Nakao A, Nguyen HD, Kenmochi N: Analysis of ribosomal protein gene structures: implications for intron evolution. PLoS genetics. 2006, 2 (3): e25-10.1371/journal.pgen.0020025.
Ohno S: Evolution by Gene Duplication. 1970, New York , Springer-Verlag
Ferrier DE, Holland PW: Ancient origin of the Hox gene cluster. Nature reviews. 2001, 2 (1): 33-38. 10.1038/35047605.
Blomme T, Vandepoele K, De Bodt S, Simillion C, Maere S, Van de Peer Y: The gain and loss of genes during 600 million years of vertebrate evolution. Genome biology. 2006, 7 (5): R43-10.1186/gb-2006-7-5-r43.
Dehal P, Boore JL: Two rounds of whole genome duplication in the ancestral vertebrate. PLoS biology. 2005, 3 (10): e314-10.1371/journal.pbio.0030314.
Jaillon O, Aury JM, Brunet F, Petit JL, Stange-Thomann N, Mauceli E, Bouneau L, Fischer C, Ozouf-Costaz C, Bernot A, Nicaud S, Jaffe D, Fisher S, Lutfalla G, Dossat C, Segurens B, Dasilva C, Salanoubat M, Levy M, Boudet N, Castellano S, Anthouard V, Jubin C, Castelli V, Katinka M, Vacherie B, Biemont C, Skalli Z, Cattolico L, Poulain J, De Berardinis V, Cruaud C, Duprat S, Brottier P, Coutanceau JP, Gouzy J, Parra G, Lardier G, Chapple C, McKernan KJ, McEwan P, Bosak S, Kellis M, Volff JN, Guigo R, Zody MC, Mesirov J, Lindblad-Toh K, Birren B, Nusbaum C, Kahn D, Robinson-Rechavi M, Laudet V, Schachter V, Quetier F, Saurin W, Scarpelli C, Wincker P, Lander ES, Weissenbach J, Roest Crollius H: Genome duplication in the teleost fish Tetraodon nigroviridis reveals the early vertebrate proto-karyotype. Nature. 2004, 431 (7011): 946-957. 10.1038/nature03025.
Liti G, Louis EJ: Yeast evolution and comparative genomics. Annual review of microbiology. 2005, 59: 135-153. 10.1146/annurev.micro.59.030804.121400.
Haber JE, Debatisse M: Gene amplification: yeast takes a turn. Cell. 2006, 125 (7): 1237-1240. 10.1016/j.cell.2006.06.012.
Narayanan V, Mieczkowski PA, Kim HM, Petes TD, Lobachev KS: The pattern of gene amplification is determined by the chromosomal location of hairpin-capped breaks. Cell. 2006, 125 (7): 1283-1296. 10.1016/j.cell.2006.04.042.
Lynch M, Conery JS: The origins of genome complexity. Science (New York, NY. 2003, 302 (5649): 1401-1404.
Pal C, Papp B, Lercher MJ: An integrated view of protein evolution. Nature reviews. 2006, 7 (5): 337-348. 10.1038/nrg1838.
Gilbert W, de Souza SJ, Long M: Origin of genes. Proceedings of the National Academy of Sciences of the United States of America. 1997, 94 (15): 7698-7703. 10.1073/pnas.94.15.7698.
Liu M, Grigoriev A: Protein domains correlate strongly with exons in multiple eukaryotic genomes--evidence of exon shuffling?. Trends Genet. 2004, 20 (9): 399-403. 10.1016/j.tig.2004.06.013.
Liu M, Walch H, Wu S, Grigoriev A: Significant expansion of exon-bordering protein domains during animal proteome evolution. Nucleic acids research. 2005, 33 (1): 95-105. 10.1093/nar/gki152.
Hille B: Ion Channels of Excitable Membranes. 2001, Sunderland, MA , Sinauer, 3rd ed.
Miller C: An overview of the potassium channel family. Genome biology. 2000, 1 (4): REVIEWS0004-10.1186/gb-2000-1-4-reviews0004.
Bichet D, Haass FA, Jan LY: Merging functional studies with structures of inward-rectifier K(+) channels. Nature reviews. 2003, 4 (12): 957-967. 10.1038/nrn1244.
Murata Y, Okado H, Katsuyama Y, Okamura Y, Kubo Y: Primary structure, developmental expression and functional properties of an inward rectifier K+ channel of the tunicate. Receptors & channels. 2001, 7 (5): 387-399.
Murata Y, Okado H, Kubo Y: Characterization of heteromultimeric G protein-coupled inwardly rectifying potassium channels of the tunicate tadpole with a unique pore property. The Journal of biological chemistry. 2001, 276 (21): 18529-18539. 10.1074/jbc.M009644200.
Okamura Y, Nishino A, Murata Y, Nakajo K, Iwasaki H, Ohtsuka Y, Tanaka-Kunishima M, Takahashi N, Hara Y, Yoshida T, Nishida M, Okado H, Watari H, Meinertzhagen IA, Satoh N, Takahashi K, Satou Y, Okada Y, Mori Y: Comprehensive analysis of the ascidian genome reveals novel insights into the molecular evolution of ion channel genes. Physiological genomics. 2005, 22 (3): 269-282. 10.1152/physiolgenomics.00229.2004.
Kuo A, Gulbis JM, Antcliff JF, Rahman T, Lowe ED, Zimmer J, Cuthbertson J, Ashcroft FM, Ezaki T, Doyle DA: Crystal structure of the potassium channel KirBac1.1 in the closed state. Science (New York, NY. 2003, 300 (5627): 1922-1926.
Derst C, Karschin C, Wischmeyer E, Hirsch JR, Preisig-Muller R, Rajan S, Engel H, Grzeschik K, Daut J, Karschin A: Genetic and functional linkage of Kir5.1 and Kir2.1 channel subunits. FEBS letters. 2001, 491 (3): 305-311. 10.1016/S0014-5793(01)02202-5.
Nishida M, MacKinnon R: Structural basis of inward rectification: cytoplasmic pore of the G protein-gated inward rectifier GIRK1 at 1.8 A resolution. Cell. 2002, 111 (7): 957-965. 10.1016/S0092-8674(02)01227-8.
Finley M, Arrabit C, Fowler C, Suen KF, Slesinger PA: betaL-betaM loop in the C-terminal domain of G protein-activated inwardly rectifying K(+) channels is important for G(betagamma) subunit activation. The Journal of physiology. 2004, 555 (Pt 3): 643-657. 10.1113/jphysiol.2003.056101.
Vibranovski MD, Sakabe NJ, de Oliveira RS, de Souza SJ: Signs of ancient and modern exon-shuffling are correlated to the distribution of ancient and modern domains along proteins. Journal of molecular evolution. 2005, 61 (3): 341-350. 10.1007/s00239-004-0318-y.
Doolittle RF: Convergent evolution: the need to be explicit. Trends in biochemical sciences. 1994, 19 (1): 15-18. 10.1016/0968-0004(94)90167-8.
Pipiras E, Coquelle A, Bieth A, Debatisse M: Interstitial deletions and intrachromosomal amplification initiated from a double-strand break targeted to a mammalian chromosome. The EMBO journal. 1998, 17 (1): 325-333. 10.1093/emboj/17.1.325.
Eichler EE, Sankoff D: Structural dynamics of eukaryotic chromosome evolution. Science (New York, NY. 2003, 301 (5634): 793-797.
Mark MD, Herlitze S: G-protein mediated gating of inward-rectifier K+ channels. European journal of biochemistry / FEBS. 2000, 267 (19): 5830-5836. 10.1046/j.1432-1327.2000.01670.x.
Connors NC, Adams ME, Froehner SC, Kofuji P: The potassium channel Kir4.1 associates with the dystrophin-glycoprotein complex via alpha-syntrophin in glia. The Journal of biological chemistry. 2004, 279 (27): 28387-28392. 10.1074/jbc.M402604200.
Doring F, Derst C, Wischmeyer E, Karschin C, Schneggenburger R, Daut J, Karschin A: The epithelial inward rectifier channel Kir7.1 displays unusual K+ permeation properties. J Neurosci. 1998, 18 (21): 8625-8636.
Hebert SC, Desir G, Giebisch G, Wang W: Molecular diversity and regulation of renal potassium channels. Physiological reviews. 2005, 85 (1): 319-371. 10.1152/physrev.00051.2003.
Ishii M, Horio Y, Tada Y, Hibino H, Inanobe A, Ito M, Yamada M, Gotow T, Uchiyama Y, Kurachi Y: Expression and clustered distribution of an inwardly rectifying potassium channel, KAB-2/Kir4.1, on mammalian retinal Muller cell membrane: their regulation by insulin and laminin signals. J Neurosci. 1997, 17 (20): 7725-7735.
Poopalasundaram S, Knott C, Shamotienko OG, Foran PG, Dolly JO, Ghiani CA, Gallo V, Wilkin GP: Glial heterogeneity in expression of the inwardly rectifying K(+) channel, Kir4.1, in adult rat CNS. Glia. 2000, 30 (4): 362-372. 10.1002/(SICI)1098-1136(200006)30:4<362::AID-GLIA50>3.0.CO;2-4.
Ikuta T, Yoshida N, Satoh N, Saiga H: Ciona intestinalis Hox gene cluster: Its dispersed structure and residual colinear expression in development. Proceedings of the National Academy of Sciences of the United States of America. 2004, 101 (42): 15118-15123. 10.1073/pnas.0401389101.
Ferrier DE, Holland PW: Ciona intestinalis ParaHox genes: evolution of Hox/ParaHox cluster integrity, developmental mode, and temporal colinearity. Molecular phylogenetics and evolution. 2002, 24 (3): 412-417. 10.1016/S1055-7903(02)00204-X.
Ford CE, Skiba NP, Bae H, Daaka Y, Reuveny E, Shekter LR, Rosal R, Weng G, Yang CS, Iyengar R, Miller RJ, Jan LY, Lefkowitz RJ, Hamm HE: Molecular basis for interactions of G protein betagamma subunits with effectors. Science (New York, NY. 1998, 280 (5367): 1271-1274.
Albsoul-Younes AM, Sternweis PM, Zhao P, Nakata H, Nakajima S, Nakajima Y, Kozasa T: Interaction sites of the G protein beta subunit with brain G protein-coupled inward rectifier K+ channel. The Journal of biological chemistry. 2001, 276 (16): 12712-12717. 10.1074/jbc.M011231200.
Mirshahi T, Mittal V, Zhang H, Linder ME, Logothetis DE: Distinct sites on G protein beta gamma subunits regulate different effector functions. The Journal of biological chemistry. 2002, 277 (39): 36345-36350. 10.1074/jbc.M205359200.
Wall MA, Coleman DE, Lee E, Iniguez-Lluhi JA, Posner BA, Gilman AG, Sprang SR: The structure of the G protein heterotrimer Gi alpha 1 beta 1 gamma 2. Cell. 1995, 83 (6): 1047-1058. 10.1016/0092-8674(95)90220-1.
Sondek J, Bohm A, Lambright DG, Hamm HE, Sigler PB: Crystal structure of a G-protein beta gamma dimer at 2.1A resolution. Nature. 1996, 379 (6563): 369-374. 10.1038/379369a0.
Pagel P, Mewes HW, Frishman D: Conservation of protein-protein interactions - lessons from ascomycota. Trends Genet. 2004, 20 (2): 72-76. 10.1016/j.tig.2003.12.007.
Fraser HB, Hirsh AE: Evolutionary rate depends on number of protein-protein interactions independently of gene expression level. BMC evolutionary biology. 2004, 4: 13-10.1186/1471-2148-4-13.
Saeed R, Deane CM: Protein protein interactions, evolutionary rate, abundance and age. BMC bioinformatics. 2006, 7: 128-10.1186/1471-2105-7-128.
Dong K, Tang L, MacGregor GG, Hebert SC: Localization of the ATP/phosphatidylinositol 4,5 diphosphate-binding site to a 39-amino acid region of the carboxyl terminus of the ATP-regulated K+ channel Kir1.1. The Journal of biological chemistry. 2002, 277 (51): 49366-49373. 10.1074/jbc.M208679200.
JGI-Ciona genome. [http://genome.jgi-psf.org/ciona4/ciona4.home.html]
Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG: The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic acids research. 1997, 25 (24): 4876-4882. 10.1093/nar/25.24.4876.
Kumar S, Tamura K, Nei M: MEGA3: Integrated software for Molecular Evolutionary Genetics Analysis and sequence alignment. Briefings in bioinformatics. 2004, 5 (2): 150-163. 10.1093/bib/5.2.150.
This work was supported in part by a grant for the promotion of the advancement of education and research in graduate schools from the Ministry of Education, Culture, Sports and Technology of Japan, and partly by a grant-in-aid for scientific research from the same Ministry to M. Tanaka-Kunishima. In addition, we appreciate the Misaki Marine Biological Station, Graduate School of Science, University of Tokyo, lending us (M. Tanaka-Kunishima and K. Takahashi) their seashore laboratory for marine biological experiments.