| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Call For Papers: Comparative Genomics
Laboratory of Bioscience, Faculty of Engineering, Iwate University, 4-3-5 Ueda, Morioka 020-8551, Japan
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
temperature-sensitive ion channel; vertebrate; comparative genomics; gene duplication; phylogenetic analyses
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
All nine known temperature-sensitive channels belong to the transient receptor potential (TRP) superfamily (5, 30, 31, 48, 49). Due to their ion channel properties, these nine channels are called "thermoTRPs." The TRP superfamily is further divided into subfamilies based on the similarity of their amino acid sequences. Among the nine thermoTRPs, four belong to the TRPV (V for vanilloid) subfamily (TRPV1, TRPV2, TRPV3, and TRPV4), four to the TRPM (M for melastatin) subfamily (TRPM2, TRPM4, TRPM5, and TRPM8), and one to the TRPA (A for ankyrin) subfamily (TRPA1). In mammals, the TRP superfamily currently consists of 
30 members that sense diverse physical and chemical stimuli (2, 5, 30, 31, 48, 49), suggesting that repeated gene duplications and subsequent functional diversifications have played fundamental roles in the evolution of TRP superfamily.
The nine thermoTRPs are activated by distinct temperature thresholds. TRPV1 is activated by temperatures >42°C, TRPV2 by >52°C, TRPV3 by >33°C, TRPV4 between 27 and 42°C, TRPM2 between 35 and 42°C, TRPM4 and TRPM5 between 15 and 35°C, TRPM8 by <25°C, and TRPA1 by <17°C, when overexpressed in cultured cells or Xenopus oocytes (7, 8, 22, 23, 30–32, 39, 40, 44, 48–51). Virtually the entire range of temperatures that mammals are exposed is covered by the temperature threshold of these nine thermo-TRPs. All thermoTRPs, except for the TRPM2, M4, and M5, are expressed in the peripheral nervous system. Therefore, six thermoTRPs (TRPV1–4, TRPM8, and TRPA1) are suggested to be involved in ambient and body temperature perception in mammals. For TRPV1, V3, and V4, knockout mice have been constructed, and behavioral responses have been examined (6, 21, 25). These mice showed deficits in responses to temperature consistent with temperature ranges of destructed thermo-TRPs. On the other hand, conflicting results have been reported for TRPA1 knockouts generated in two different laboratories (4, 20). One group claimed that TRPA1 knockout mice show deficit in response to noxious cold (20), while the other group did not observe such effects on cold temperature perception (4). In addition, such unequivocal observations have been reported when TRPA1 was overexpressed in cultured cells (17, 40). Therefore, the role for TRPA1 in cold temperature perception has been disputed.
Some of the thermoTRPs are expressed in tissues other than neuron such as skin in the case of TRPV1, V3, and V4 (10, 30, 33, 51). It has been suggested that keratinocytes may act as thermal receptors since synapses have not been found between keratinocytes and sensory termini. Putative signal transduction from keratinocytes has been proposed to be mediated by an ATP-gated channel (P2X3) that is present in sensory termini (30, 33). TRPM5 is expressed in the taste buds of the tongue and has been suggested to function downstream of the taste receptors and to be involved in the modulation of thermal sensitivity in sweet taste perception (41).
Interestingly, thermoTRPs has been reported to be activated by stimuli other than temperature. For example, TRPV1 is also activated by acidic pH and capsaicin, TRPV2 by growth factor molecules, TRPV4 by hypotonic stimulus, TRPM8 by menthol and icilin, and TRPA1 by icilin (2, 8, 22, 23, 30–32, 40, 48). Therefore, it appears that thermoTRPs serve not only as thermosensors but can also transmit other kinds of sensations.
Mammalian thermoTRP homologs have recently been identified in invertebrates. For example, TRPA1 has been reported in sea squirt (Ciona intestinalis), fruit fly (Drosophila melanogaster), and nematode (Caenorhabditis elegans) (29, 49). In the case of D. melanogaster, TRPA1 has been shown to be activated by warmth (approximately >27°C, Ref. 45) and to be involved in thermotaxis (35). These findings indicate that TRPA1 arose and was involved in a thermosensation mechanism in an early stage of animal evolution. As mammalian TRPA1 is activated by cold (<17°C; Refs. 40, 45) rather than warmth, its activation temperature threshold must have changed in the course of animal evolution. On the other hand, mammalian orthologous genes for TRPV1–4 and TRPM2, M4, M5, M8 have not been found in C. intestinalis, D. melanogaster, and C. elegans genome sequences (29). These observations suggest that these thermoTRPs arose after the divergence of the urochordate and vertebrate lineages. However, the identification of very few thermoTRP homologs in nonmammalian species to date has impeded elucidation of the evolutionary processes of thermoTRPs in vertebrate lineages. Recently, however, the draft genome sequences have been released for several key model organisms from different vertebrate classes and have provided an opportunity for the genome-wide screening for thermoTRP homologs.
In the present study, we conducted comprehensive searches for the genes of mammalian thermoTRP homologs in the draft genome sequences of chicken (Gallus gallus), western clawed frog (Xenopus tropicalis, diploid species closely related to the tetraploid species Xenopus laevis), zebrafish (Danio rerio), and pufferfish (Fugu rubripes). Together with well-characterized thermoTRPs of humans and rodents, the thermoTRP homologs were compared among species belonging to the four major different vertebrate classes: mammals, birds, amphibians, and fishes. Phylogenetic analyses of thermoTRP homologs were then carried out with the aim of elucidating the evolutionary processes of thermoTRPs in vertebrates. Since the thermosensor repertoire directly affects thermosensation mechanism, comparative analyses of thermoTRP homologs supply basic information for understanding the evolutionary changes of the thermosensation mechanisms in vertebrate lineages.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
600-amino-acid portion of the NH2-terminal were excluded since these genes do not possess the central transmembrane domains that are well conserved in the TRP ion channels (Supplementary Fig. 1). (The online version of this article contains supplemental data.) We thus obtained 36 predicted sequences from the assembly annotations of the respective draft genome sequences. These predicted mRNA sequences were aligned with previously reported thermoTRPs whose full-length cDNA sequences were available in the databases.
|
|
|
In the present study, we used gene names according to the unified nomenclature for the TRP superfamily (24). The names of newly identified genes were assigned according to the names of the known genes contained in the same clusters in the phylogenetic tree. The thermoTRP homologs used in the present study are listed in Table 1.
Molecular phylogenetic and evolutionary analyses.
The amino acid sequences were deduced by translating mRNA sequences. Multiple sequence alignments were performed using the CLUSTAL W algorithm (12), with minor manual adjustments. Evolutionary distances between the amino acid or nucleotide sequences were calculated with the Jones-Taylor-Thornton (JTT) (16) or Tamura-Nei (42) models, respectively, after all alignment gap sites were eliminated. From the evolutionary distances thus determined, phylogenetic trees were reconstructed by the neighbor-joining (NJ) or minimum-evolution (ME) methods (36, 37). The statistical confidence of each branch in the phylogenetic tree was estimated by the bootstrap method (11) with 1,000 replications. The rates of nonsynonymous (dN) and synonymous (dS) substitutions, and their ratios (dN/dS) were estimated for all pairs of vertebrate TRPV4s by the modified Nei-Gojobori method (52) with the transition/transversion ratio equal to 1.5 and Jukes-Cantor correction (18). The standard error was estimated by the bootstrap method with 1,000 replications. For testing the purifying selection, Z-tests were performed under the null hypothesis of dS = dN. All of the data analyses were performed using MEGA 3 (19).
Synteny detection.
To identify the orthologous genes surrounding the thermoTRP genes, we performed best-hit gene analyses. For each comparison, the gene of one species (species 1) located in the vicinity of the thermoTRP gene was used as query, and a TBLASTN search was first performed against the genome sequence of another species (species 2). When the best-hit gene in the genome sequence of species 2 was located around the orthologous thermoTRP gene, we used the amino acid sequence of this gene as a query and performed a TBLASTN search against the genome sequence of species 1. If the best-hit gene in the second TBLASTN search was the same gene used as the query for the first TBLASTN search, we defined this gene pair as orthologous genes. To further examine whether the conserved synteny exists around the thermoTRP orthologs between different species, we subsequently performed the alignments of the genome sequences between the regions containing the thermoTRP orthologs using the PipMaker (38). When the results obtained by the PipMaker were consistent with those obtained by best hit gene analysis, we considered it a confirmation of our results.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
TRPV subfamily in vertebrates.
We identified four copies of mammalian thermoTRP homologs in chicken, nine in western clawed frog, two in zebrafish, and three in pufferfish (Table 1). The amino acid sequences of these TRPVs were aligned with those of TRPVs available in the databases (Supplementary Fig. 3), and the phylogenetic trees were reconstructed using conserved portion containing predicted ANK repeat, transmembrane (TM), pore loop (PL), and TRP domains (Supplementary Figs. 1 and 3). A phylogenetic tree reconstructed from amino acid sequences using the ME method is shown in Fig. 1. The vertebrate TRPVs were divided into four major clusters that were supported by 100% bootstrap values. The clusters are designated as TRPV1/2, TRPV3, TRPV4, and TRPV5/6 according to the known TRPs included in the respective clusters. The newly predicted genes identified in the present study were named according to the known genes contained in the respective clusters. Although TRPV5s and TRPV6s are not thermo-TRPs, we included these genes in the phylogenetic analysis; inclusion of TRPV5s and TRPV6s in the analysis enables them to be used as out-groups since they diverged first from the three remaining clusters.
|
|
|
Phylogenetic analysis of the vertebrates TRPV subfamily revealed that the chicken possesses one copy each of TRPV1, V2, V3, and V4, that the western clawed frog possesses one copy each of TRPV1, V2, and V3, and six copies of TRPV4, that the zebrafish possesses one copy each of TRPV1/2 and V4, and that the pufferfish possesses two copies of TRPV1/2 and one copy of TRPV4 (Fig. 1). The numbers of thermoTRPV homologs in each species are summarized in Table 3. Note that TRPV5 and TRPV6 are not thermoTRP and thus not included in the table.
|
36.4 Mb away from each other. Therefore, TRPV4s and the other TRPVs were located at different genomic locations in all species examined. The gene orders around TRPV4 among human, chicken, and zebrafish were well conserved (data not shown). In the genome sequence of the western clawed frog, TRPV4a and TRPV4b–f were located in scaffold 17 and 330, respectively. The gene order around TRPV4a and other vertebrate TRPV4s were well conserved (data not shown). TRPV4b–f were located within an 167-kb region and organized tandemly without intervening genes in the following order TRPV4b, V4c, V4d, V4e, and V4f. Several genes were located at each side of the TRPV4a and TRPV4b–f in the respective scaffolds, thus TRPV4a was not organized in tandem with TRPV4b–f in the genome sequence of western clawed frog.
TRPV1, V2, and V3 were located close to each other. Figure 3 shows the gene orders surrounding TRPV1, V2, and V3 among the human, chicken, western clawed frog, and zebrafish. In the genome sequences of the human, chicken, and western clawed frog, TRPV1 and TRPV3 were located adjacently. Furthermore, in the chicken and western clawed frog, TRPV2 was located near TRPV1 and TRPV3 (within 0.58 Mb). On the other hand, in human chromosome 17, TRPV2 was located 12 Mb away from TRPV1 and TRPV3, further away from the latter two TRPVs. The gene orders around TRPVs were well conserved among the human, chicken, and western clawed frog. The gene order around TRPV1/2 of zebrafish was also conserved, although we could not find TRPV2 and TRPV3 orthologs in the fish genomes. The pufferfish TRPV1/2a and TRPV1/2b were located in scaffold 511 and 188, respectively, thus located in different genomic regions (Supplementary Fig. 5A). The gene orders around pufferfish TRPV1/2a and zebrafish TRPV1/2 were conserved (data not shown), suggesting these two genes are orthologs as suggested by phylogenetic analysis (Fig. 1).
|
97% bootstrap values. The clusters were designated as TRPM2, TRPM4, TRPM5, and TRPM8 according to the names of the known genes contained in the respective clusters. TRPM2 and TRPM8 clusters diverged first from remaining two clusters. All four clusters contained tetrapod and fish TRPMs, except that the TRPM8 cluster did not contain fish TRPM8s. The TRPM8 cluster contained two genes of western clawed frog that formed a monophyletic cluster. The extent of the divergence between these two copies corresponded to that of the divergence between TRPM8s of chicken and mammals. TRPM8a and TRPM8b of the western clawed frog were arranged tandemly in scaffold 15, suggesting that these genes were produced by tandem gene duplication event.
|
The numbers of vertebrate thermoTRPM homologs are summarized in Table 3. All of the vertebrate TRPM homologs examined in the present study possessed six predicted TM domains and one TRP domain (Supplementary Figs. 1 and 6).
Several TRP channels (TRPM8, TRPV1, TRPV3) have recently been reported to exhibit a shift in the voltage dependence of their channel activations (shift their current-voltage relationships) in response to temperature stimuli; and this trait has been proposed to be coupled with the temperature sensitivity of thermoTRPs (27, 47). Although the voltage sensor has not yet been identified in thermoTRPs, the TM4 domain may be a good candidate since, in the case of the Shaker K+ channel, positively charged amino acids in the TM4 domain were shown to act as a voltage sensor (1). In this respect, it is worth noting that the positively charged arginine residue located in the predicted TM4 domain of the TRPMs and the TRPVs was entirely conserved (Supplementary Figs. 3 and 6).
TRPA1 homologs of vertebrates.
The searches for TRPA1 homologs identified one copy each in the chicken, western clawed frog, and pufferfish, and two in zebrafish (Table 1). The amino acid sequences of these TRPA1s were aligned with those of TRPA1 of mammals, sea squirt (C. intestinalis), insects (D. melanogaster and A. gambiae), and nematode (C. elegans) (Supplementary Fig. 7). Phylogenetic trees were then reconstructed from the amino acid sequences by the NJ and ME methods (Fig. 5). The topologies of the phylogenetic trees inferred by both methods were consistent. The phylogenetic relationship of vertebrate TRPA1s was consistent with the established speciation process of vertebrates. Two copies of zebrafish TRPA1s did not form a monophyletic cluster, and zebrafish TRPA1b clustered with pufferfish TRPA1. Zebrafish TRPA1a and TRPA1b were located in chromosome 2 and 24, respectively. It is worth noting that sea squirt TRPA1 clustered with insect TRPA1s with high bootstrap values (93% and 86% in the NJ and ME trees, respectively) consistent with Okamura et al.'s findings (29), although the sea squirt is considered to be more closely related to vertebrates than to insects. All of the vertebrate TRPA1s possessed the predicted 15 ANK repeat domains and six TM domains (Supplementary Figs. 1 and 7). The numbers of vertebrate TRPA 1 homologs are summarized in Table 3.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Using the thermoTRP homologs detected in the present study, we performed phylogenetic analyses to elucidate evolutionary processes in vertebrate lineages. In the phylogenetic tree of the TRPV subfamily, the TRPV4 cluster was located at a position ancestral to the TRPV3 cluster (Fig. 1). Comparison of the genomic locations of the TRPVs revealed that TRPV4 was located at a different genomic location in tetrapods and fish, although TRPV1, V2, and V3 were closely located (Fig. 3). This arrangement is consistent with the hypothesis that TRPV4s diverged before the divergence of TRPV1, V2, and V3.
Although the TRPV4 cluster contained tetrapod and fish genes, the TRPV3 cluster did not contain fish genes. In addition, both the ME and NJ trees strongly indicate that fish TRPV1/2s were produced after the divergence between TRPV3 and TRPV1/2 clusters (Fig. 1 and Supplementary Fig. 4A). Therefore, we conclude that TRPV3 emerged in a common ancestor of fishes and tetrapods and has been lost in fish lineages. The inferred evolutionary processes of thermoTRP homologs of vertebrates are shown in Fig. 6.
The time of the gene duplication event producing TRPV1 and TRPV2 could not be unambiguously determined. In the NJ tree inferred from amino acid sequences, the gene duplication event appeared to occur after the divergence between fish and tetrapod lineages (Supplementary Fig. 4A). In the ME tree, however, the fish TRPV1/2s clustered with tetrapod TRPV1s (Fig. 1), suggesting that TRPV1 and TRPV2 already existed in the common ancestor of fishes and tetrapods and that TRPV2 was subsequently lost in the fish lineage. When the genomic location of zebrafish TRPV1/2 was examined, its genomic position corresponded to that of tetrapod TRPV1 (Fig. 3). In the genome sequences of the chicken and western clawed frog, TRPV1 and TRPV3 were located adjacently, whereas TRPV2 was located several genes away from TRPV1 and TRPV3. Taking into consideration the findings that the gene orders around the tetrapod TRPV2 were also conserved near the zebrafish TRPV1/2, we find it likely that three TRPVs may have already existed in the common ancestor of fishes and tetrapods and that they were organized in tandem, TRPV2-TRPV1-TRPV3, in the genome sequence. If so, then several genes were later inserted between TRPV1 and TRPV2 before the divergence of fish and tetrapod lineages, and TRPV2 and TRPV3 were subsequently lost in the common ancestor of zebrafish and pufferfish. Accordingly, the genomic organization of TRPVs is consistent with the phylogenetic relationship reconstructed by the ME method.
In the phylogenetic tree of the TRPM subfamily, all clusters except for the TRPM8 cluster contained both tetrapod and fish genes (Fig. 4). This situation suggests that the gene duplication events producing TRPM2, M4, M5, and M8 occurred before the divergence of fish and tetrapod lineages and that TRPM8s were lost in the common ancestor of zebrafish and pufferfish (Fig. 6). Meanwhile our findings indicate that TRPM4 and TRPM5 have been lost in the chicken and western clawed frog, respectively. Since TRPM4 and TRPM5 are reported to exhibit similar ion channel properties (41), they might be able to compensate for each other's functions if either were lost. The zebrafish possesses three copies of TRPM4s (termed a, b, and c). Amino acid sequence similarity and close proximity of the genomic location of TRPM4b and TRPM4c suggest they were produced by a tandem gene duplication event. On the other hands, TRPM4a did not cluster together with TRPM4b and TRPM4c but, rather, clustered with pufferfish TRPM4 (Fig. 4), suggesting that the gene duplication event occurred in the common ancestor of zebrafish and pufferfish and that one copy was lost in the pufferfish lineage (Fig. 6).
With regard to fish lineages, the timing of gene duplication events appeared to be similar for TRPV1/2, TRPM4, and TRPA1. Specifically, these gene duplication events occurred just before the divergence of zebrafish and pufferfish lineages (Figs. 1, 4, and 5). In this respect, it has been proposed that the whole genome duplication event occurred in the common ancestor of zebrafish and pufferfish and greatly affected the fish genome structures (9, 43). To examine whether three gene duplications of thermoTRPs were associated with genome duplication event, we compared the gene orders around each pair of the duplicated copies of TRPV1/2, TRPM4, and TRPA1. We found that several genes located adjacent to the duplicated copies of TRPs showed high amino acid sequence similarity to one another (from 47 to 88%, Supplementary Fig. 5) in the case of TRPV1/2s and TRPM4s, although a similar relationship was not found in the case of TRPA1a and TRPA1b. These findings suggest that duplicated copies of TRPV1/2s and TRPM4s are located in paralogous genomic regions. Therefore, it is likely that TRPV1/2s and TRPM4s were produced by the whole genome duplication event occurring in the common ancestor of zebrafish and pufferfish.
Although gene duplication events have occurred in the fish lineage, fishes have also subsequently lost TRPV1/2, TRPM4, and TRPA1. In the phylogenetic trees of the TRPV, TRPM, and TRPA1 subfamilies, the two duplicated copies from the same species did not form monophyletic clusters (Figs. 1, 4, and 5), suggesting that gene loss occurred after the gene duplication events (Fig. 6). Furthermore, pufferfish TRPV1/2b has deletion in the predicted TM domains; thus this TRPV may be a pseudogene, or at least it may possess the different ion channel properties (Supplementary Fig. 3). In addition, TRPV2, TRPV3, and TRPM8 were lost in the common ancestor of zebrafish and pufferfish (Fig. 6). Therefore, in the fish lineages, the thermoTRP homologs seem to have a tendency to be lost. If the activation temperature thresholds of fish thermoTRPs are similar to those of mammalian thermoTRPs, the activation temperature ranges of three thermoTRPs that fishes possess (TRPV1/2, TRPV4, and TRPA1) could roughly cover the entire range of ambient water temperature (30, 31, 48). In this respect, fishes may possess reduced sets of thermoTRP channels compared with those of tetrapods. Alternatively, other as yet uncharacterized channels may serve as the thermosensors in fishes; if so fishes would possess a different thermoTRP repertoire from mammals. Hence, examination of the activation temperature thresholds of fish thermoTRP homologs is an interesting subject for future investigation.
In the present study, we found that most of the thermoTRPs that mammals possess already existed in the common ancestor of tetrapods and ray-finned fishes and that thermoTRP homolog repertoires have changed in the lineages leading to different vertebrate classes (Table 3 and Fig. 6). The absence of orthologous genes of mammalian thermoTRPs beyond TRPA1 in urochordates (sea squirts) (29) suggests that most of the vertebrate thermoTRPs emerged after the divergence of vertebrate and urochordate lineages but before that of ray- and lobe-finned fish lineages. In this respect, identification and comparison of the thermoTRP homologs in organisms that diverged in the early stage of vertebrate evolution, such as cartilaginous fishes (sharks), and jawless fishes (lampreys), and Cephalochorda (amphioxuses), will help us to better understand of the origin of vertebrate thermoTRPs.
Implications for functional evolution of vertebrate thermo-TRPs.
The only mammalian thermoTRP ortholog known to be present in sea squirt (C. intestinalis) is TRPA1 (29), and TRPA1 orthologs exist in fruit fly (D. melanogaster) and nematode (C. elegans). Thus it can be concluded that TRPA1 emerged quite early in animal evolution. The finding that C. intestinalis TRPA1 is more similar to insect TRPA1 than to vertebrate TRPA1s is inconsistent with the speciation process as reported by Okamura et al. (29; see also Fig. 5). The activation temperature thresholds of TRPA1s are different between D. melanogaster and human when they were overexpressed in Xenopus oocytes (45). The functional portions responsible for this interspecies difference are not yet known. Examination of the temperature threshold of C. intestinalis TRPA1 may supply insight into change in activation temperature threshold of TRPA1 in the course of animal evolution.
Our phylogenetic analysis of vertebrate thermoTRP homologs revealed that all of the thermoTRPs that mammals possess already existed in the common ancestor of tetrapods (Fig. 6), indicating that the genetic basis for thermosensation of warm-blooded animals, in terms of their thermosensor repertoires, was acquired in the common ancestor of tetrapods. Acquisition of homeothermism occurred after the divergence of amphibians and bird/mammal lineages and may have significantly affected the thermosensation mechanism. Since amphibians diverged just before the acquisition of homeothermism characteristic of mammals and birds, comparisons of thermoTRPs between amphibians and warm-blooded animal (birds and mammals) are valuable for better understanding the mechanism of thermosensation. In this respect, it is worth noting that the NH2- and COOH-terminal portions of the predicted amino acid sequences of TRPV3 of the western clawed frog were truncated compared with those of mammalian TRPV3s (Supplementary Fig. 3). Mammalian TRPV3 channels are activated by temperatures >33°C and are involved in sensing temperatures around the body temperature of warm-blooded animals (25, 33, 39, 51).
Since the predicted amino acid sequence of TRPV3 of the western clawed frog was not supported by the ESTs, prediction of the NH2 and COOH-terminal portions may not be correct. Therefore, we performed TBLASTN searches with the E value set <10–5 against the draft genome sequences of the western clawed frog using the amino acid sequences of the NH2 and COOH-terminal portions of the mammalian and chicken TRPV3s as queries; however, no hits were obtained. These results suggest that the NH2 and COOH-terminal portions of TRPV3 of the western clawed frog differ from those of TRPV3s of mammals and birds. Truncations of the COOH-terminal portion of TRPV1 reduced its temperature threshold when expressed in cultured cells (46), thus the COOH-terminal portion is suggested to be responsible for determining the temperature thresholds of TRPV channels. Determination of the cDNA sequences of amphibian TRPV3s and examination of their temperature thresholds may provide intriguing insights into the evolution of thermosensory function relevant to the evolutionary acquisition of homeothermism of mammals and birds.
Interestingly, western clawed frog not only possesses homologs for most of the mammalian thermoTRPs but has also gained and diversified TRPM8 and TRPV4 homologs (Figs. 2, 4 and 6). TRPM8a and TRPM8b, as well as TRPV4b–f, are located in tandem in the western clawed frog genome sequence, respectively; thus tandem duplication is the main mechanism that produced diversified copies of TRPV4 and TRPM8 homologs.
Phylogenetic analysis and comparisons of the dN/dS ratios among vertebrate TRPV4s revealed that the functional constraint of TRPV4a is similar to that of other vertebrate TRPV4s, which have been reported to be activated by temperature and hypotonic stimuli in in vitro expression experiments (22, 50) and thus are considered to be functional channels (Table 2). The other five copies (TRPV4b–f) have diverged considerably, and their dN/dS ratios suggest that substantial relaxation of the functional constraint has occurred in these lineages. However, in most of the comparisons, dSs are significantly larger than dNs, suggesting that most, if not all, of the TRPV4s have been subjected to purifying selection. In addition, all of the predicted amino acid sequences of TRPV4s in the western clawed frog possessed conserved structural domains, and EST data suggest that the mRNAs of TRPV4a, TRPV4b, and TRPV4d genes are transcribed. These observations suggest that diversified copies of TRPV4s of the western clawed frog maintain their function. After the divergence of TRPV4a and the ancestral gene of TRPV4b–f, considerable amino acid substitutions accumulated in the ancestral gene of TRPV4b–f (Fig. 2), which may contribute to the acquisition of different properties, while TRPV4a retained its original function. Subsequently, the ancestral gene of TRPV4b–f gained copies through repeated tandem gene duplications, and further accumulations of the amino acid substitutions in each of the copies led to the diversified copies of TRPV4s in the western clawed frog.
Taking into consideration the fact that other vertebrate species possess only one copy of TRPV4, TRPV4s of the western clawed frog may have conferred some advantages that might be related to the traits specific to amphibians. TRPV4s of mammals and chickens are activated not only by temperature but also by hypotonic stimuli and are thought to serve as an osmosensor (22, 28). As amphibians have a unique subterrestrial life cycle that require them to be well adapted to both aquatic and terrestrial environments, the western clawed frog may utilize diversified copies of TRPV4s as osmosensors to accurately sense the osmotic pressure of its skin. Further examination of the functional properties of TRPV4s in amphibians is essential for understanding the functional evolution of TRPV4s.
In summary, in the present study we found that the repertoires of thermoTRP homologs have changed through vertebrate evolution. Since thermoTRPs are involved in thermosensation as well as other kinds of sensory detection (2, 30, 31, 48), variability of repertoires may be associated with adaptation of the organisms to their respective habitat environments. Further examination of the functions of thermoTRP homologs in organisms other than mammals will enhance our understanding of how the functional changes of thermoTRPs have contributed to the vertebrate adaptation mechanisms.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
ali A, Hudspeth AJ, Friedman JM. Vanilloid receptor-related osmotically activated channel (VR-OAC), a candidate vertebrate osmoreceptor. Cell 103: 525–523, 2000.[CrossRef][ISI][Medline]
ánková K, Lyfenko A, Ettrich R, Vyklicky L. Functional role of C-terminal cytoplasmic tail of rat vanilloid receptor 1. J Neurosci 23: 1340–1350, 2003.This article has been cited by other articles:
|
|
 |
|
  D. A. Prober, S. Zimmerman, B. R. Myers, B. M. McDermott Jr, S.-H. Kim, S. Caron, J. Rihel, L. Solnica-Krezel, D. Julius, A. J. Hudspeth, et al. Zebrafish TRPA1 Channels Are Required for Chemosensation But Not for Thermosensation or Mechanosensory Hair Cell Function J. Neurosci., October 1, 2008; 28(40): 10102 - 10110. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 |
