Functional characterization of thermosensitive TRPV channels from holocephalan elephant shark (Callorhinchus milii) illuminate the ancestral thermosensory system in vertebrates

Animals have evolved a physiological system for sensing ambient and body temperature in order to respond to changes in environmental temperature that impact survival. Species adapted to different thermal niches may have acquired specific temperature sensitivities appropriate to their habitats. Some transient receptor potential (TRP) channels are temperature sensitive and function as sensory molecules for thermoperception in a variety of animals (Himmel and Cox, 2020; Hoffstaetter et al., 2018). TRPs are tetrameric, non-selective cation channels with six transmembrane domains located in the plasma membrane. They have been identified in a wide range of species from yeast to humans (Clapham, 2003). In mammals, 11 temperature-sensitive TRP (thermoTRP) channels have been reported, belonging to the subfamilies named vanilloid (V), melastatin (M), ankyrin (A) and canonical (C). The respective TRP channels sense different temperature ranges to cover physiologically relevant temperature ranges in mammals. Some TRP channels sense extreme heat or cold and serve as nociceptive receptors (Hoffstaetter et al., 2018).

Many thermoTRP channels have multimodal properties and are activated by non-thermal stimuli such as osmotic pressure, mechanical stimuli and various chemical substances (Saito and Tominaga, 2017). For example, mouse and human TRPV1 are activated by heat (43°C or higher), acid (proton) and chemicals such as capsaicin, a component of chili peppers (Caterina et al., 1997). Similarly, in mammals, TRPV3 is activated by warmth (∼33–39°C) close to their body temperature, as well as camphor, eucalyptol (components of eucalyptus) and 2-aminoethoxy diphenyl borate (2-APB) (Peier et al., 2002b; Smith et al., 2002; Xu et al., 2002; Caterina, 2014). TRPV4 is also activated by temperatures of 27–34°C or higher, low osmotic pressure and mechanical stimulation (Caterina, 2014).

TRP orthologs differ in their sensitivity to temperature and chemicals depending on the species (Saito and Tominaga, 2015). TRPV1 of the tropical clawed frog Xenopus tropicalis is activated by high temperature (∼40°C or higher), acid and capsaicin. However, it is 1000-times less sensitive to capsaicin than the rat ortholog (Ohkita et al., 2012; Saito et al., 2016). Zebrafish TRPV1 acts as a molecular sensor of environmental temperature. It has been reported to be activated at temperatures higher than ∼25°C (Gau et al., 2013), whereas another study showed activation at 33°C (Gracheva et al., 2011). TRPV1 of masu salmon (Oncorhynchus masou ishikawae) is also heat activated with a threshold near 28°C (Yoshimura et al., 2022). Xenopus tropicalis TRPV3 is activated by 2-APB but not by camphor or eucalyptol. High temperatures also did not stimulate X. tropicalis TRPV3, whereas a low temperature of ∼16°C activated X. tropicalis TRPV3 (Saito et al., 2011). In contrast, the TRPV3 homolog in human is activated at temperatures higher than approximately 39°C (Smith et al., 2002). TRPV4 is identified from several fishes including the European sea bass, medaka, zebrafish, flatfish and tilapia (Mangos et al., 2007; Bossus et al., 2011; Amato et al., 2012; Sánchez-Ramos et al., 2012; Seale et al., 2012; Hori et al., 2022). The detailed channel properties have been investigated in medaka. It has been reported that medaka TRPV4 exhibits both heat and cold sensitivities with a threshold of 13°C and 40°C, respectively (Hori et al., 2022).

In the context of molecular evolution, TRP channels evolved through gene duplication and loss events, which produced variability in the gene repertoires among species. In general, it has been found that many TRP families were found to be shared not only among vertebrates, but also among several invertebrates (Peng et al., 2015). In vertebrate species, phylogenetic analysis revealed that most thermoTRP channels arose in a common ancestor between Sarcopterygii and Actinopterygii (Saito and Shingai, 2006; Peng et al., 2015; Matsuura et al., 2009), suggesting that the thermosensory system has been rearranged in ancestral vertebrates. However, the detailed evolutionary process of thermoTRP channels in the ancestral vertebrates remains to be discovered owing to the lack of information on species that diverged early in the evolution of vertebrate lineages.

The elephant shark Callorhinchus milii is a cartilaginous fish of the order Chondrichoda. Cartilaginous fishes are the earliest diverged taxon among jawed vertebrates. They are essential for understanding the origin and complex evolution of the physiological system in jawed vertebrates. Elephant sharks live in relatively cool environments at depths of 200 to 500 m in the ocean around southern Australia and near New Zealand, and seasonally migrate to shallower waters to lay eggs (Lyon et al., 2011). In this study, we elucidate the evolution of the thermosensory system in vertebrates by cloning and characterizing elephant shark TRPV1, TRPV3 and TRPV4 channels. Functional characterization of thermoTRP channels in the elephant shark would shed light on the evolution of thermosensory systems in ancestral vertebrates.

Both sexes of elephant shark Callorhinchus milii Bory de Saint-Vincent 1823 (Chimaeriformes) were collected from Western Port Bay, Victoria, Australia. Animals were maintained in a tank with running seawater for one week before sampling. Animals were anesthetized in 0.1% (w/v) 3-aminobenzoic acid ethyl ester. After decapitation, tissues were dissected, immediately frozen in liquid nitrogen and kept at −80°C. All animal handling and experimental procedures were approved by the Animal Care and Use Committees of the University of Tokyo.

Capsaicin (Cap), GSK1016790A (GSK), and eucalyptol were purchased from Sigma (St Louis, MO). All chemicals were dissolved in dimethylsulfoxide (DMSO) or ethanol as stock solutions for the two-electrode voltage clamp assay.

For the cloning of elephant shark TRPVs, BLAST was used to search the elephant shark genome database (https://asia.ensembl.org/Callorhinchus_milii/Info/Index?db=core) with human TRPVs as queries. Three predicted sequences corresponding to TRPV1, TRPV3 and TRPV4 were identified, and then PCR primers were designed to isolate full-length elephant shark TRPVs (TRPV1: 5′-CGTCAGCTAAAGACAAGAGATCAAGAG-3′ and 5′-CTAGACATCTTTTGGAGTCACCGAGA-3′, TRPV3: 5′-ACACGCACTTGTGAAACTTTTGGAAC-3′ and 5′-TGACAAAAGCACGAGTAAAACACTGG-3′, TRPV4: 5′-GAGCGGTCTGAAAGCAAAGCAACAG-3′ and 5′-GGTTTGATTTAATGCCAGGTTTCAG-3′). First-strand cDNA was synthesized using SuperScriptII reverse transcriptase (Invitrogen, Thermo Fisher Scientific, Waltham, MA) from 2 µg total RNA isolated from the ovary and testis. The amplified DNA fragments were subcloned with pCR-Blunt II-TOPO vector (Invitrogen), sequenced using a BigDye terminator Cycle Sequencing-kit (Applied Biosystems, Thermo Fisher Scientific) with T7 and SP6 primers, and analyzed on the 3–130 Genetic Analyzer (Applied Biosystems, Thermo Fisher Scientific). For TRPV3 channels, three alleles were found. Allele-1 has R507, N523 and R709; allele-2 has R507, K532 and R709; and allele-3 has L507, N532 and K709. We speculate that these amino acid variations may be due to individual differences.

The amino acid sequences of TRPVs for phylogenetic analysis were collected using protein–protein Blast search with a blastp option with the GenBank database (Altschul et al., 2005) (Table S1). Alignments were prepared using MUSCLE implemented in the MEGA7 program using the default parameters with amino acid sequences of several vertebrate species (Saitou and Nei, 1987; Edgar, 2004; Kumar et al., 2016). Prior to phylogenetic reconstruction, a model-fitting analysis was performed to find the model of protein evolution that best fit the dataset. The best model was the Jones–Taylor–Thornton (JTT) (Jones et al., 1992). Maximum likelihood analysis was conducted using the JTT matrix-based method (Felsenstein, 1981; Jones et al., 1992). Statistical confidence for each branch in the tree was evaluated by the bootstrap method with 1000 replications (Felsenstein, 1985).

Full-coding regions of elephant shark TRPV1, TRPV3 and TRPV4 were amplified by PCR with KOD DNA polymerase (Toyobo, Osaka, Japan). Then PCR products were gel-purified and ligated into a pOX(+) vector for oocyte expression (Dowland et al., 2000). All constructs were verified by sequencing analysis as described above.

Elephant shark TRPV1, TRPV3 or TRPV4 were heterologously expressed in oocytes of Xenopus laevis and ionic currents were recorded using the two-electrode voltage-clamp method (Saito et al., 2011). Each complementary RNA (cRNA) of elephant shark TRPVs was synthesized using pOX(+) vectors containing each of elephant shark TRPVs with an mMESSAGE MACHINE SP6 kit (Ambion, Thermo Fisher Scientific) according to the manufacturer's protocol. Fifty nanoliters of elephant shark TRPV cRNA (7.5–15 ng µl⁻¹ for TRPV1, 100–300 ng µl⁻¹ for TRPV3, 10 ng µl⁻¹ for TRPV4) was injected into defolliculated frog oocytes, and ionic currents were recorded for 1–6 days post-injection. The oocytes were voltage-clamped at −60 mV. The chemicals tested were diluted into ND96 solution (96 mmol l⁻¹ NaCl, 2 mmol l⁻¹ KCl, 1.8 mmol l⁻¹ CaCl₂, 1 mmol l⁻¹ MgCl₂ and 5 mmol l⁻¹ HEPES, pH 7.6) and applied to the oocytes by perfusion. For thermal stimulation, heated or cooled ND96 solution was applied to the oocytes. The temperature of perfused bath solutions was monitored with a thermistor located just beside the oocytes using a TC-344B (Warner Instruments, Hamden, CT). Apparent temperature thresholds for activation of TRPVs were determined with an Arrhenius plot that was generated using clampfit 10.4 (Molecular Devices, San Jose, CA) and origin 9J (OriginLab, Northampton, MA) software. Temperature thresholds for activation of TRPV1 and TRPV4 were estimated from current data obtained from X. laevis oocytes prepared from three frogs. In each preparation, data were collected from two to eight independent oocytes prepared from a single frog. Temperature thresholds for activation of TRPV3 were estimated from current data obtained from X. laevis oocytes prepared from three frogs. In each preparation, data were collected from one to seven oocytes from a single frog. All procedures involving the care and use of animals were approved by the National Institute for Physiological Sciences.

We downloaded RNA-seq reads of elephant shark for the brain, gills, heart, intestine, kidney, liver, muscle, ovary, spleen, testis and thymus from the National Center for Biotechnology Information (accession number SRP013772), and reference genome assembly and gene annotation were downloaded from Ensembl (accession ID GCA_000165045.2). RNA-seq reads of the 11 tissues were independently aligned to the reference sequences using RSEMv1.3.3 (Li and Dewey, 2011), which uses Bowtie2v2.4.1/STARv2.7.3a for alignment (Langmead and Salzberg, 2012; Dobin et al., 2013). Relative measurements of transcript abundances are expressed as FPKM (fragments per kilobase of exon per million mapped reads).

The living jawed vertebrates are represented by two lineages: the bony fishes (Osteichthyes) and cartilaginous fishes (Chondrichthyes). By their phylogenetic position, cartilaginous fishes are important for understanding the origins of complex developmental and physiological systems of jawed vertebrates (Cappetta et al., 1993; Venkatesh et al., 2014). We isolated elephant shark TRPV1, TRPV3 and TRPV4 cDNAs containing an open reading frame encoding 847, 751 and 883 amino acids, respectively (GenBank accession XM_007896535 for TRPV1, XM_007896537 for TRPV3, and XM_007901853 for TRPV4) based on the GenBank database. The overall identities of elephant shark TRPV1 with human, chicken, Xenopus tropicalis or zebrafish TRPV1 were 48, 49, 47 or 41%, respectively. The overall identities of elephant shark TRPV3 with human, chicken, X. tropicalis or whale shark TRPV3 were 43, 44, 41 and 56%, respectively. The overall identities of elephant shark TRPV4 with human, chicken, X. tropicalis or zebrafish TRPV4 were 71, 72, 72 and 67%, respectively.

TRPV sequences of selected fishes and terrestrial vertebrates (Table S1) were collected and analyzed to investigate the phylogenetic relationships of vertebrate TRPV orthologs (Fig. 1 and Fig. S1). The elephant shark TRPV1 formed a sister group with the teleost fish TRPV1, which was inconsistent with the species relationship. The clade containing elephant shark TRPV1 and teleost fish TRPV1 formed a cluster with tetrapod TRPV1, suggesting that gene duplication producing TRPV1 and TRPV2 occurred in the common ancestor of jawed vertebrates and TRPV2 has subsequently been lost in teleost and cartilaginous fish. However, the bootstrap value for the interior branch connecting tetrapod and fish TRPV1 was quite low (17%). Therefore, the timing of the gene duplication producing TRPV1 and TRPV2 could not be estimated precisely. Within the TRPV3 or TRPV4 cluster, cartilaginous fish orthologs diverged earliest from the gene of the remaining species, which was consistent with species relationships among vertebrates. The finding that cartilaginous fish possess TRPV3 suggests that it was explicitly lost in teleost fish lineages.

We heterologously expressed shark TRPV1 in X. laevis oocytes and characterized its channel property using a two-electrode voltage-clamp method. It is a conventional assay system that has been established and is widely used, and this methodology is comparable to the values that are published in the literature (Cens and Charnet, 2007; Brown et al., 2008). Since TRPV1 channels from humans, rats and clawed frogs had been reported to be activated by temperatures over 40°C (Caterina et al., 1997; Ohkita et al., 2012; Saito et al., 2016, 2019), we examined heat responses of elephant shark TRPV1. In our preliminary observations, some X. laevis oocytes expressing elephant shark TRPV1 elicited the heat-evoked currents just after the onset of heat stimulation from room temperatures (data not shown). To ensure the precise assessment of the heat-evoked response of X. laevis oocytes expressing elephant shark TRPV1, we lowered the temperature to ∼15°C before heat stimulation (Fig. 2A). Elephant shark TRPV1-expressing oocytes responded to heat stimulation (Fig. 2A,C), whereas no current was induced by heat in water-injected control oocytes (Fig. S2A). The temperature threshold for heat activation of elephant shark TRPV1 was estimated using an Arrhenius plot. Elephant shark TRPV1 showed a response to heat with a threshold of 32.0°C (Q₁₀=54.4) (Fig. 2B). The average temperature threshold for activation for elephant shark TRPV1 was 31.7±0.9°C (n=16; Fig. 2C and Table S2).

The sensitivity of TRPV1 to capsaicin varies among vertebrate species. Human, rodent and dog TRPV1 are highly sensitive to capsaicin, while orthologous channels from several species including rabbit, chicken, X. tropicalis and zebrafish show lower sensitivity to capsaicin (Saito and Tominaga, 2015). To examine the effect of capsaicin on elephant shark TRPV1, we applied 100 µmol l⁻¹ capsaicin (the concentration close to the maximum solubility in the ND96 bath solution) before or after heat stimulation. The application of capsaicin did not elicit any observable currents in the oocytes expressing elephant shark TRPV1, although they exhibited clear responses to heat (Fig. 2D,E). These results indicate that elephant shark TRPV1 is not sensitive to capsaicin at the concentrations tested (100 µmol l⁻¹), suggesting that it has no or low sensitivity to capsaicin. We also investigated the tissue distribution of TRPV1 of the elephant shark. We compared relative mRNA abundances among various tissues utilizing previously published RNA-seq data (Venkatesh et al., 2014). Elephant shark TRPV1 transcript was highly abundant in the gills and spleen (Fig. 2F).

As mammalian TRPV3 is activated by heat (33–39°C) (Xu et al., 2002; Peier et al., 2002a), we examined the effect of heat stimulation on elephant shark TRPV3. Repeated heat stimulations induced clear inward currents in oocytes expressing elephant shark TRPV3 (Fig. 3A), whereas no current was induced by heat stimulation in water-injected control oocytes (Fig. S2B). In the second heat stimulation, currents started to increase at lower temperatures than the first stimulation (Fig. 3B,E,F). The Q₁₀ values during the first heat activation were quite high but decreased extensively in the second heat activation (Fig. 3D,E). Arrhenius plots revealed that the average temperature threshold for activation at the first heat stimulation was 42.0±0.3°C (n=10; Fig. 3C,E, and Table S2). The average temperature threshold for activation at the second heat stimulation was reduced to 28.1±0.6°C (n=10; Fig. 3D,F, and Table S2). There was a significant difference in temperature thresholds for activation of elephant shark TRPV3 between the first and second heat-evoked currents (paired t-test, P=4.8×10⁻⁹). These results indicated that elephant shark TRPV3 became sensitized by repeated heat stimulation similarly to mammalian orthologous channels (Macikova et al., 2019). Eucalyptol (1,8-cineol) is a known agonist for mammalian TRPV3 (Sherkheli et al., 2009). Oocytes expressing elephant shark TRPV3 responded to the application of 5 mmol l⁻¹ eucalyptol and heat (Fig. 3G). The negative controls, water-injected oocytes, showed no response to the application of eucalyptol (Fig. S2C). These results suggested that elephant shark TRPV3 possesses channel properties that are similar to mammalian orthologs. We found elephant shark TRPV3 transcript was highly abundant in the gills and spleen and exhibited similar tissue distribution as TRPV1 (Fig. 3H).

Mammalian TRPV4 is known to be activated by warm temperature (27–34°C) (Shibasaki, 2016) and a synthetic compound, GSK1016790A (Vincent and Duncton, 2011). Applying GSK1016790A induced clear inward currents in oocytes expressing elephant shark TRPV4, whereas prior heat stimulation did not evoke any responses (Fig. 4A). The current evoked by GSK106790A stimulation was sustained even after its washout. In several thermoTRP channels, the synergistic effects between chemical and thermal stimulation were reported (Tominaga et al., 1998; Bandell et al., 2004). Therefore, we assumed that heat might have positively affect elephant shark TRPV4. Unexpectedly, the application of heat suppressed sustained current evoked by GSK1016790A (Fig. 4A). These observations are similar to the phenomenon found in cold-activated TRPM8, where currents evoked by menthol (an agonist) stimulation were suppressed by heat (Peier et al., 2002b). Thus, we then applied cold stimulation to naïve oocytes expressing elephant shark TRPV4. Surprisingly, elephant shark TRPV4 was activated by cooling stimulation with a threshold of 20.8°C (Q₁₀=4143.8) (Fig. 4C). We found that elephant shark TRPV4 is activated by gradually lowering the temperature from room temperature (24°C) (Fig. 4D). The average thermal activation thresholds were 20.4±0.38°C (n=10; Fig. 4D and Table S2). The cooling-evoked current of elephant shark TRPV4 exhibited a clear desensitization and nearly returned to the basal level during the low-temperature application. The average temperature at peak currents elicited by cooling stimulation was 14.7±0.48°C (n=10; Fig. 4D and Table S2). The factors causing desensitization of channel activity (temperature or duration of stimulation) remain to be resolved. Neither cold nor GSK1016790A stimulation induced inward currents in water-injected oocytes (Fig. 4E). These results indicated that elephant shark TRPV4 is a cooling-activated channel. Furthermore, gene expression analysis using RNA-seq data revealed that elephant shark TRPV4 transcript is highly expressed in the liver (Fig. 4F).

Comparative analyses of TRP channels from different species provide essential information on their structure–function relationships and the molecular evolution of TRP. Here, we performed the molecular cloning and characterization of the orthologs of TRPV1, TRPV3 and TRPV4 from the holocephalan elephant shark Callorhinchus milii, which diverged at an early stage of the jawed vertebrate lineage. Elephant shark TRPV1 was activated by heat stimulation in our heterologous expression system using Xenopus oocytes, indicating that its thermal sensitivity is preserved even in cartilaginous fish. The thermal activation threshold of elephant shark TRPV1 (31.7±0.9°C) was similar to the thresholds reported for several other fish species, but was much lower than those of mammalian orthologs (Gau et al., 2013; Yoshimura et al., 2022; York, 2023). TRPV1 from human, rodent and clawed frog were reported to be activated by temperatures higher than 40°C (Caterina et al., 1997; Peier et al., 2002b; Ohkita et al., 2012). It is worth noting that elephant sharks generally inhabit the southern coast of Australia and around New Zealand at depths of 200–500 m on the continental shelf (Venkatesh et al., 2014). From spring to summer, they migrate to shallow waters such as bays and estuaries to spawn during the breeding season (Hyodo et al., 2007; Malcolm, 1997). Therefore, elephant sharks are generally thought to live in cool environments. However, elephant sharks, their embryos and fry may be exposed to relatively high temperatures when migrating to shallow waters, especially during the breeding season, as the sea surface temperature in the habitat area is ∼20°C (Boisvert et al., 2015). TRPV1 functions as a pain receptor in a variety of vertebrates, including humans, rodents and frogs, but the TRPV1 channel, which activates at about 30°C, appears to be an ancestral property of fish (Gau et al., 2013; Yoshimura et al., 2022; York, 2023).

We could not observe capsaicin-evoked responses in elephant shark TRPV1. Previous studies identified three amino acid residues involved in the binding of capsaicin to TRPV1: serine, threonine, and glutamic acid at the position 512 (S512), 550 (T550) and 570 (E570) around the transmembrane region 3 and 4 of rat TRPV1, respectively (Fig. S2D; Jordt and Julius, 2002; Gavva et al., 2004; Saito and Tominaga, 2015; Saito et al., 2016; Chu et al., 2020). We examined three corresponding amino acid residues in elephant shark TRPV1 and found that cysteine replaced the amino acid residue corresponding to S512 in rat TRPV1. However, Xenopus laevis TRPV1, which has the same amino acid residue at the same position, is sensitive to capsaicin, suggesting that this amino acid substitution might not be responsible for the lower sensitivity of elephant shark TRPV1 to capsaicin. In contrast, amino acid substitution at the corresponding position at 550 residue in rat TRPV1 might be involved since this amino acid was isoleucine in elephant shark TRPV1. This amino acid is the same in rabbit and zebrafish TRPV1, which show lower sensitivity to capsaicin (Fig. S2D). In addition, the amino acid corresponding to E570 in rat TRPV1 was substituted to glutamine in zebrafish and elephant shark TRPV1, which might be related to their reduced sensitivity to capsaicin.

Our previous phylogenetic analyses of the TRPV subfamily showed that TRPV3 emerged in the common ancestor of vertebrates and was subsequently lost in teleost fish lineages. However, the detailed evolutionary process of TRPV3 remained elusive (Saito and Shingai, 2006; Saito et al., 2011). The identification of TRPV3 in the elephant shark and whale shark strongly suggests that it was acquired at least in the common ancestor of jawed vertebrates, as the loss of TRPV3 was specific to teleost fishes.

Oocytes injected with elephant shark TRPV3 responded to heat stimulation and to eucalyptole, a known TRPV3 agonist. The temperature activation threshold of elephant shark TRPV3 at the first heat stimulation was approximately 42°C, which was higher than that of mammalian TRPV3 (approximately 33 to 39°C). Although rodent TRPV3 was previously reported to be activated by warmth, recent studies showed that it is activated by extreme heat (50°C) in the initial activation. Subsequently, its thermal activation threshold is reduced in the subsequent heat stimulation (Liu and Qin, 2017). In the present study, we also observed a significant decrease in the thermal activation threshold of elephant shark TRPV3 in the second heat stimulation. Therefore, elephant sharks and mammals roughly share the property of TRPV3, suggesting that it evolved as a heat- and chemical-sensitive channel in the common ancestor of jawed vertebrates.

Unexpected findings were obtained by characterizing elephant shark TRPV4. TRPV4 has been reported to be activated by warm temperatures and the synthetic agonist GSK1015790A in mammals and reptiles (Benemei et al., 2015; Lawhorn et al., 2020; Yatsu et al., 2015). However, elephant shark TRPV4 showed no response to heat, even though it was activated by GSK1016790A. Instead, the GSK1016790A-evoked activity of elephant shark TRPV4 was suppressed by heat. The suppression of chemically (GSK101679A)-evoked currents by heat was similar to the observations found in cold-activated TRPM8 (Myers et al., 2009). These observations support our idea that elephant shark TRPV4 is a bona fide channel with sensitivity to cooling. The opposite thermal sensitivity among orthologous channels has been previously reported for TRPV3 between mammals and frogs (Saito et al., 2011). Furthermore, mammalian and teleost fish TRPA1s have been reported to be activated by both heat and cold (Sinica et al., 2019; Oda et al., 2016). Recently, medaka TRPV4 was also reported to be activated by both heat and cold stimulation (Hori et al., 2022). Taken together with previous findings, our results indicate that TRPV4 may have evolved as a cool (or bimodal) thermal sensor in the ancestors of jawed vertebrates, and may subsequently change its thermal sensitivity in tetrapods in response to environmental changes in tetrapod and bony fish lineages, respectively.

Our present results suggest that elephant shark TRPV4 is activated in response to decreases in temperature, with an average thermal activation threshold of ∼20°C. During the breeding season, elephant sharks migrate to shallow water where temperatures exceed 20°C, suggesting that TRPV4 may function as a thermal sensor in such circumstances. However, the physiological role of TRPV4 in deep-sea environments, where temperatures range from 12 to 15°C, remains unclear. Further characterization of TRPV4 in elephant sharks and other vertebrates may provide clues to understand the physiological role of TRPV4 in fishes.

The mRNA abundance of elephant shark TRPV1 and TRPV3 was high in the gills and spleen, and TRPV4 mRNA was found at high levels in the liver. Recently, Flores-Aldama et al. (2020) reported that elephant shark TRPV5/6 showed high expression in gills. TRPV1 has also been reported to be expressed in the gills of masu salmon (Yoshimura et al., 2022). Gills that are in contact with the external environment and exposed to fluctuation in osmolarity and temperatures. Thus, elephant shark TRPV1 and TRPV3 might be highly expressed in gills to sense ambient water temperatures accurately. In murine, TRPV4 is expressed in cholangiocytes, representing a mechanosensor responsible for translating fluid flow into intracellular signaling and biliary secretion (Li et al., 2020). Still, the functional role of TRPV4 in the liver of the elephant shark is unclear. Detailed analyses of the expression level and localization of TRPV4 are necessary to understand its functional role in the liver of the elephant shark.

In this study, we isolated and functionally characterized elephant shark TRPV1, TRPV3 and TRPV4. This report is the first complete sequence information and characterization of TRPVs in Chondrichthyes, cartilaginous fishes that are basal to lobe-and ray-finned fishes. Previous phylogenetic analyses revealed that the gene repertoire of thermosensitive TRP channels was rearranged in the ancestral vertebrates. However, it was unclear whether newly emerged TRP channels acquired thermal sensitivity just after their birth. Here, we found that all three elephant shark TRPV channels are thermosensitive, suggesting that they evolved as thermal sensors at least 470 million years ago in the common ancestor of jawed vertebrates.

Although all three elephant shark TRP channels characterized in our study exhibited thermal responses, further investigations are needed to determine whether they play a pivotal role in thermal sensation in elephant sharks because our analysis was conducted using a heterologous expression system. Notably, the thermal activation thresholds of TRPV1 and TRPV3 exceed the temperature ranges of the deep sea, where adult elephant sharks typically live. These findings suggest the possibility that TRPV1 and TRPV3 may not function as thermal sensors in vivo. Alternatively, their thermal sensitivities might be modulated under physiological conditions, allowing activation within temperature ranges relevant to elephant sharks. For example, thermal sensitivity of TRPA1 in Antarctic fishes has been shown to change under specific cellular conditions (York, 2023). In summary, this study refines the evolutionary process of the thermosensory systems during the early stage of vertebrate evolution. Furthermore, our data provide a valuable approach for future studies investigating the basic physiology of cartilaginous fish TRPs.

We are indebted to Dr Lina C. Wang, Minnesota State University, Mankato for a critical reading of this manuscript. We thank colleagues in our laboratories. The authors thank the technical staff of Hokkaido University and National Institute for Physiological Sciences for expert technical assistance throughout the study.