Adrenomedullin (AM), known as a multifunctional hormone in mammals, forms a unique family of five paralogous peptides in teleost fish. To examine their cardiovascular effects using homologous AMs in eels, we isolated cDNAs encoding four eel AMs, and named AM1 (ortholog of mammalian AM), AM2, AM3 (paralog of AM2 generated only in teleost lineage), and AM5 according to the known teleost AM sequences. Unlike pufferfish, not only AM1 but AM2/3 and AM5 were expressed ubiquitously in various eel tissues. Synthetic mature AM1, AM2, and AM5 exhibited vasodepressor effects after intra-arterial injections, and the effects were more potent at dorsal aorta than at ventral aorta. This indicates that AMs preferentially act on peripheral resistance vessels rather than on branchial arterioles. The potency was in the order of AM2 = AM5 ≫ AM1 in both freshwater (FW) and seawater (SW) eels, which is different from the result of mammals in which AM1 is as potent as, or more potent than, AM2 when injected peripherally. The minimum effective dose of AM2 and AM5 in eels was 1/10 that of AM1 in mammals. The hypotension reached 50% at 1.0 nmol/kg of AM2 and AM5, which is much greater than atrial natriuretic peptide (20%), another potent vasodepressor hormone. Even with such hypotension, AMs did not change heart rate in eels. In addition, AM1 increased blood pressure at ventral aorta and dorsal aorta immediately after an initial hypotension at 5.0 nmol/kg, but not with AM2 and AM5. These data strongly suggest that specific receptors for AM2 and AM5 exist in eels, which differ from the AM1 receptors identified in mammals.
- calcitonin gene-related peptide
- ventral aorta
- dorsal aorta
- vasodepressor action
homeostasis of the cardiovascular system is maintained by the regulation of various cardiovascular parameters such as blood pressure, blood volume, and heart rate. In fishes, the regulation of blood pressure and flow at the osmoregulatory organs is of particular importance for cardiovascular regulation. For example, blood delivery to the arterio-arterial and the arterio-venous circulation greatly affect ion and water exchange in the gills and thus blood volume (13, 46). Further, blood pressure at renal glomeruli of the teleost kidney profoundly modulate glomerular filtration rate, which is a major determinant for urine flow rate in teleost fish (4). Blood pressure and flow in peripheral circulation, as well as local circulation in the gills and the kidney, are under neural, endocrine, and local control by paracrine factors. Among such controls, endocrine control appears to play integral roles, and several hormones such as angiotensin (ANG), arginine vasotocin (AVT), endothelins, natriuretic peptides, bradykinin, and urotensins have been implicated in cardiovascular regulation through their potent actions on the cardiovascular system, as demonstrated both in vivo and in vitro (47). In addition, adrenomedullin (AM) is a new candidate for such important regulators of the cardiovascular system in fishes (5).
AM is a relatively new biologically active peptide isolated from human pheochromocytoma cells of adrenal medulla origin (20). AM is recognized as a member of calcitonin gene-related peptide (CGRP) family that consists of CGRP, AM, and amylin. AM exerts potent vasodepressor actions in mammals, mainly through the vasorelaxant effect on peripheral resistant vessels (10, 25, 32, 53). The gene expression of AM has been detected in various organs and tissues, including neural, cardiovascular, endocrine, renal, and immune tissues (6, 21, 22, 30, 35). AM seems to have multiple functions at these organs and tissues, including body fluid regulation (51), immunomodulation (1), secretion of other hormones (19), and other functions, in addition to cardiovascular regulation. In nonmammalian species, the presence of immunoreactive AM has been reported in fishes, amphibians, reptiles, and birds (34), and a cDNA encoding AM has been isolated in birds and teleost fishes (23, 41, 66). It is noteworthy that five AMs, named AM1 to AM5, exist and comprise an independent subfamily in several teleost species (41, 59). Using a comparative genomic approach, Ogoshi et al. (42) suggested a history of diversification of the whole CGRP family in vertebrates. They further showed that teleost AMs can be divided into three groups, AM1/4, AM2/3, and AM5, and that mammalian AM is an ortholog of teleost AM1 (41). On the basis of these findings, they identified AM2 and AM5 in selected species of mammals (42, 57). AM2 is also named intermedin (50) and thus usually described as AM2/IMD (60). However, AM2 was used here to distinguish clearly from other teleost AMs.
Consistent with the presence of multiple AMs in teleost fish, AM receptors seem to have diversified in this vertebrate group. In mammals, members of the CGRP family bind receptors consisting of calcitonin receptor or calcitonin receptor-like receptor (CLR) coupled with one of the three receptor activity-modifying proteins (RAMPs) identified in mammals, and the binding increases intracellular cAMP production (37). AM1 induces cAMP accumulation in cells transfected with a combination of CLR and RAMP2 or RAMP3, and AM2 was effective to CLR-RAMP3 combination with a low potency (50, 59). On the other hand, Nag et al. (40) identified three CLRs (CLR1 to CLR3) and five RAMPs (RAMP1 to RAMP5), in mefugu, Takifugu obscurus. They showed that AM1 accumulated cAMP in cells transfected with combinations of CLR1-RAMP2/3/5 and CLR2-RAMP2, but AM2 and AM5 were effective only with a combination of CLR1-RAMP3. Furthermore, the activity of AM1 for cAMP accumulation in the CLR1-RAMP3 was greater than that of AM2 or AM5. These findings indicate that multiple AMs and their receptor systems (including unknown receptors) may have novel functions in fishes, but little is known about the physiological functions of AMs in this vertebrate group.
The physiological actions of AM have been examined intensively in mammals, especially on the cardiovascular, body fluid, and immune regulation (1, 5, 19, 51). Similar effects of newly identified AM2 have also been examined in mammals (9, 15, 18, 57, 63). However, no investigation has been performed on the physiological functions of multiple AMs in teleost fishes using homologous system. The aim of this study is to examine the cardiovascular effect of homologous AMs using eels (Anguilla japonica), in which methods for cardiovascular experiments in conscious animals, including cannulation into various blood vessels, have been established. Thus far, however, multiple AMs have not been identified in teleost species other than those in which genome or EST databases are available, such as pufferfish, medaka, zebrafish, and salmonids (41, 59). Therefore, AM cDNAs were initially cloned in eels, and their phylogenetical analyses and tissue distribution were examined for identification of AM species. Then putative mature sequences of eel AM1, AM2, and AM5, which belong to three different groups of the AM subfamily, were chemically synthesized, and injected into the dorsal aorta of conscious eels in fresh water (FW) or seawater (SW) to examine their effects on dorsal and ventral aortic blood pressure and heart rate.
MATERIALS AND METHODS
Cultured eels, Anguilla japonica (body wt 180.2 ± 3.6 g, n = 23), were purchased from a local dealer. They were acclimated in a FW or SW tank (50 liters) without feeding for more than a week (FW eels) and 2 wk (SW eels) before use. Water in the tank was filtrated, aerated, and maintained at 18°C. All animal experiments described in this paper were approved by the Animal Experiment Committee of the University of Tokyo and performed in accordance with the Manual for Animal Experiments.
After anesthesia in 0.1% (wt/vol) tricaine methanesulfonate (Sigma, St. Louis, MO, USA), the brain, pituitary, gill, heart, interrenal, head kidney, kidney, esophagus, stomach, intestine, liver, pancreas, spleen, mesentery, red body, skin, and skeletal muscle were isolated from eels and immediately frozen in liquid nitrogen. Total RNA was extracted using ISOGEN (Nippongene, Toyama, Japan) for subsequent experiments.
cDNA cloning of eel AMs.
Double-strand cDNA pool was prepared from 1.0 μg of total RNA using SMART cDNA Library Construction kit (Clontech Labratories, Palo Alto, CA, USA). The partial regions of four AMs (AM1, AM2, AM3, and AM5) were amplified using degenerate primers that were designed according to the known nucleotide sequences of each AM in teleost fish (Table 1). Gene-specific primers were designed on the basis of the partial sequences, and then the 5′ or 3′ region of the cDNAs were amplified by 5′ rapid amplification of cDNA ends (RACE) or 3′ RACE method using 5′ PCR primer or CDS III/3′ primer in the kit and the gene-specific primers (Table 1). Finally, the cDNAs that encompass the whole coding regions of eel AMs were amplified using gene-specific primers designed on the basis of the partial sequences determined in RACE method. Amplified products were subcloned into pT7blue vector (Novagen, Madison, WI, USA) and then sequenced by a 310 DNA sequencer (PerkinElmer, Yokohama, Japan).
Phylogenetic analyses of eel AMs.
Putative amino acid sequences of cloned eel AM precursors were aligned with those of other species using the ClustalX version 1.83 (ftp://ftp-igbmc.u-strasbg.fr/pub/ClustalX/), which was followed by manual adjustments of inconsistencies. A phylogenetic tree was constructed by Bayesian method, using MrBayes ver. 3.1.2 software (http://mrbayes.csit.fsu.edu/).
Tissue distribution of AM mRNA.
Tissue distribution of AM transcripts was examined by RT-PCR. Single-strand cDNAs of various tissues were prepared from 1.0 μg of total RNA using Superscript First Strand Synthesis System (Invitrogen, Carlsbad, CA), and PCR was performed using Ex Taq DNA polymerase (TAKARA). The β-actin cDNA (accession no. AB074846) was used for an internal standard. The PCR was performed as follows: initial denaturation at 94°C for 3 min was followed by 25 cycles for AM1 and β-actin, 30 cycles for AM2, 27 cycles for AM3, and 28 cycles for AM5 of denaturation (94°C for 30 s), annealing (60°C, 30 s for AM1 and AM3, 60°C, 1 min for AM5 and β-actin, and 65°C, 1 min for AM2) and extension (72°C, 30 s for AMs and 1 min for β-actin). The amplified DNA fragments were electrophoresed on a 1.2% agarose gel and detected by ethidium bromide staining.
Predicted mature peptides of eel AM1, AM2, AM5, and atrial natriuretic peptide (ANP) were chemically synthesized by the Peptide Institute (Osaka, Japan). These peptides were synthesized by a peptide synthesizer (430A; Applied Biosystems, Foster City, CA, USA) with p-methyl-benzhydrylamine resin as a solid support. The correct sequence was confirmed by mass analysis, amino acid analysis, and reverse-phase HPLC. The COOH termini of eel AMs were amidated as deduced from other AMs. Among a few type of AM2, amino acid sequences as shown in Fig. 2 were synthesized and used in physiological study. Human AM1 and AM2 were purchased from the Peptide Institute. Each peptide was dissolved in distilled water at a concentration of 10−4 M, aliquoted and kept at −20°C until use. The stock solution was diluted with 0.9% NaCl containing 0.01% Triton-100.
Twelve FW eels (185.6 ± 4.2 g) and five SW eels (180.4 ± 7.6 g) were used for the experiments. Eels were anesthetized in 0.1% (wt/vol) tricaine methanesulfonate for 15 min and placed on an operation board for surgery. The ventral and dorsal aortas were cannulated with polyethylene tubes (0.8 mm OD) for bolus injections of hormones and measurement of blood pressure, as described previously (55). After surgery, eels were placed in a plastic trough through which aerated water circulated at 18°C. On the next day of surgery, the tubes were connected via a three-way stopcock to pressure transducers (Nihon Koden, Tokyo, Japan) for continuous monitoring of blood pressure at the ventral aorta (PVA) and dorsal aorta (PDA). The signal was amplified by a carrier amplifier (7903; NEC San-ei, Tokyo, Japan), and recorded by a pen recorder (Rikadenki, Tokyo, Japan). To measure PVA and PDA, five eels were injected with 0.1 and 1.0 nmol/kg of eel AM1, AM2, and AM5 in 0.05 ml of 0.9% NaCl into the dorsal aorta in a random order of doses. As the vasodepressor effects were more potent at dorsal aorta than at ventral aorta, it was evaluated whether the vasodepressor effects were dose dependent by measurement of PDA using seven FW eels. FW eels were injected with 0.01, 0.1, and 1.0 nmol/kg of eel AM1, AM2, AM5, human AM2, or eel ANP in 0.05 ml of 0.9% NaCl into the dorsal aorta in random order of doses. Each injection was immediately followed by an injection of 0.05 ml of saline to flush the dead volume of tubes (0.08 ml). Injection intervals were placed more than 1 h to ensure reproducible response. The cardiovascular responses were evaluated in terms of maximal changes after injection of hormones. As performed in FW eels, the cardiovascular responses were evaluated in SW eels at the dose of 0.005, 0.05, 0.5, and 5.0 nmol/kg of eel AM1, AM2, and AM5.
The changes in arterial blood pressure were expressed in terms of the percentage of maximal change to the arterial blood pressure compared with the preinjection level. The difference between ventral aorta and dorsal aorta was compared by Mann-Whitney's U-test. Decrease in PDA after injection of each peptide was compared with saline by Wilcoxon test. Decreases in PDA and changes of heart rate were compared among eel AMs by Steel-Dwass test at each dose and time point. The vasodepressor effects of each AM were compared with that of ANP by Steel test at each dose. Significance was determined at P < 0.05. All results were expressed as means ± SE.
cDNA cloning of eel AMs.
Four cDNAs encoding eel AMs were amplified from the RNA of eel brain or kidney (Fig. 1). These genes were identified as orthologs of AM1, AM2, AM3, and AM5 on the basis of the structural characteristics of other AM cDNAs and on the homology of putative mature sequence of cloned eel AM with those of other animals sequenced to date (Fig. 2). Eel AM1 encoded 173 amino acid residues, and the putative mature sequence was connected with the COOH-terminal peptides by four consecutive arginine (Arg) residues, which may be removed by carboxypeptidase for the COOH-terminal amidation using a glycine (Gly) residue, as observed in mammalian AMs. A proadrenomedullin N-terminal 20 peptide (PAMP)-like sequence at the NH2 terminus of prohormone, which is characteristic to AM1, also existed in eel AM1. Eel AM2 and AM3 consisted of 175 and 165 amino acid residues, respectively, and the COOH-terminal Gly residue was followed by a stop codon. Eel AM5 consisted of 124 amino acid residues, and the COOH-terminal Gly residue was followed by an Arg residue prior to a stop codon as is the case with teleost AM5's. Single nucleotide polymorphisms were detected in the AM2 cDNA that changed amino acid length and sequence in the putative mature peptide and in the AM1 and AM5 cDNAs that altered amino acid sequence in the prosegment (data not shown). Eel AMs had common structural characteristics, such as an amidation signal at COOH terminus of putative mature peptide and the possible intramolecular ring structure formed by a disulfide bond. Amino acid sequences of eel AM2 and AM3 were highly conserved between fishes and mammals compared with AM1 or AM5 (Fig. 2). A cDNA coding for eel AM4, a paralog of AM1 that is generated at the third round whole genome duplication in the teleost lineage (42), could not be isolated in this study, probably because of large sequence variability of this peptide.
The phylogenetic analyses of the precursor protein confirmed that four eel AMs were classified into each group of AMs (Fig. 3). However, eel AM2 and AM3 were placed at the juxtaposition in phylogenetic tree and not classified with pufferfish AM2 and AM3, respectively. This may be accounted for by unusually conserved sequences of AM2 and AM3 in teleost fish. The nucleotide sequences of eel AMs have been registered to the DDBJ/EMBL/GenBank nucleotide sequence databases (accession nos. AB363985 to AB363988).
Tissue distribution of AM mRNA.
Expression of the eel AM genes was detected in various eel tissues when examined by RT-PCR (Fig. 4). AM1 mRNA was strongly expressed in the heart, kidney, liver, spleen, and red body. AM2 and AM3 mRNA were ubiquitously detected in various tissues, particularly abundant in the pituitary and spleen (AM2) and in the kidney, esophagus, liver, and spleen (AM3). AM5 mRNA was also ubiquitously expressed, particularly in the spleen and red body. In the spleen, all eel AMs were strongly expressed. There were little differences in the tissue distribution and the expression level of four eels AMs between FW and SW-adapted eels.
Cardiovascular effects of AMs.
Mean PVA and PDA measured before injections were 29.4 ± 2.7 (n = 5) and 19.4 ± 1.6 mmHg (n = 12) in FW eels, respectively. All eel AMs at more than 0.1 nmol/kg consistently depressed PVA and PDA in a similar time course, which reached the lowest level in 5 min after injection (Fig. 5). The decrease in PDA was more profound than that of PVA, and the difference was significant at 1.0 nmol/kg of eel AM5 in FW eels (Figs. 5 and 6A). The vasodepressor effects of eel AMs were dose dependent at doses of 0.01, 0.1, and 1.0 nmol/kg (Fig. 6B). Bolus injections of eel AM1 at 0.01, 0.1, and 1.0 nmol/kg depressed PDA by 2.0 ± 2.3%, −7.7 ± 1.4% and −13.4 ± 2.3% (n = 7), respectively. The decrease of PDA was more profound by eel AM2 and AM5 than by eel AM1: −11.3 ± 2.1%, −20.6 ± 2.4%, and −52.9 ± 3.9% by eel AM2, and −10.4 ± 2.2%, −23.2 ± 4.1%, and −50.0 ± 3.2% (n = 7 in each case) by AM5 at 0.01, 0.1, and 1.0 nmol/kg, respectively (Fig. 6B). These vasodepressor effects were significant compared with saline injection (P < 0.05), except for AM1 at 0.01 nmol/kg. The effect of eel AM2 and AM5 was more potent and efficacious than that of eel AM1 but equipotent to heterologous, human AM2 (Fig. 6B, Table 2). In contrast to human AM2, heterologous human AM(1) had no effect on PDA even at a dose of 1.0 nmol/kg. Eel ANP, which is another potent vasodepressor hormone in eels, decreased PDA to the same degree as eel AM2 and AM5 at a dose of 0.01 and 0.1 nmol/kg, but the decrease by 1.0 nmol/kg of AM2 and AM5 was significant compared with that of ANP (P < 0.01) (Fig. 6B, Table 2). Even with such profound hypotension after injection of AM2 and AM5, heart rate did not change significantly despite a tendency to increase, except for those at 15 min after injections of AM1 and AM5, and at 20 min after injection of AM1 (Fig. 7).
Mean PVA and PDA in SW eels measured before injections were 28.6 ± 1.7 mmHg (n = 5) and 20.0 ± 0.4 mmHg (n = 5), respectively. Bolus injections of eel AMs to SW eels also depressed arterial blood pressure in a dose-dependent manner with a potency order of AM2 = AM5 ≫ AM1 (Fig. 8). The statistical analysis did not detect significant differences in the potency between FW and SW eels. The vasodepressor effects of AMs were usually greater in PDA than in PVA as observed in FW eels (Fig. 8). Interestingly, after injection at 5.0 nmol/kg of AM1, the small pressure decrease was always followed by an immediate, profound increase in arterial pressure (Fig. 9A). The hypertension reached maximum within 10 min after injection. The initial decrease in PDA was significantly greater than in PVA, but the subsequent hypertension was comparable between the two aortas (Fig. 9B). Such secondary hypertension was not observed at 5.0 nmol/kg of AM2 and AM5.
In this study, we have shown that eel AM2 and AM5 are two of the most potent and efficacious vasodepressor hormones thus far examined in teleost fish. The teleost AM peptides comprise a unique family of five paralogous peptides, which can be divided into three groups, AM1/4, AM2/3, and AM5 (41, 42). However, no investigation has been performed on their physiological actions, except for a preliminary report by Aota (3), using human AM showing that bolus injections at 3.0 and 7.0 nmol/kg depressed dorsal aortic pressure in the rainbow trout but not in the cod. It is likely that the low vasodepressor activity may be due to the use of heterologous mammalian AM that may have low sequence identity (51% between eel and human) with teleost AM1 (Fig. 2). Therefore, this is the first report that examined the cardiovascular effects of three major members of the AM family (AM1, AM2, and AM5) in nonmammalian species using homologous peptides. On the basis of the fish studies, AM2/IMD and AM5 have recently been identified in selected species of mammals (42, 50, 57). In nonmammalian species other than teleost fish, AM1 has been identified in birds (66).
Initially, we attempted to examine the cardiovascular effects in the pufferfish (T. rubripes), in which five AMs have been identified (41). However, cannulation into both ventral and dorsal aorta was impossible in this species. Therefore, this study was performed using eels, because in vivo techniques for measurements of cardiovascular parameters have been established in this species. To use homologous AM peptides in eels, we initially attempted to identify all five AMs by cDNA cloning, but attempts to isolate AM4 cDNA were not successful, probably because of low sequence identity of AM4 in teleost species (59). Ogoshi et al. (42) showed by comparative genomic analyses that AM1, AM2, and AM5 had existed before divergence of actinopterygian and sarcopterygian (tetrapod) lineages, and they are duplicated at the third round whole genome duplication that occurred only in the teleost lineage. After duplication, duplicated AM4 changed considerably, AM3 changed only slightly, and AM5 counterpart seems to have disappeared. On the basis of the evolutionary history of AM peptides, we decided to synthesize eel AM1, AM2, and AM5 for physiological experiments that examine their cardiovascular actions.
The phylogenetic analysis of the precursor largely classified cloned four eel AMs into each cluster (Fig. 3). However, as classification of eel AM2 and AM3 was not straightforward, as their mature peptides were extremely similar to each other in eels (90%), AM2 and AM3 were named on the basis of the sequence identity with Takifugu AM2 and AM3. Compared with Takifugu AMs, eel AM2 had higher sequence identity with Takifugu AM2 (83.1%) than Takifugu AM3 (82.8%), and eel AM3 had higher sequence identity with Takifugu AM3 (84.2%) than Takifugu AM3 (81.8%). Eel AM1 prohormone has a PAMP-like sequence at the NH2 terminus and a COOH terminus peptide that is connected with the mature peptide by consecutive Arg residues as do other AM1 prohormones identified thus far (35). In mammals, it is known that the AM1 gene produces mRNA that codes for only PAMP by alternative splicing (36), but such a product was not detected in eels as confirmed by RT-PCR. Eel AM2 and AM3 prohormones are followed by a stop codon, and eel AM5 prohormone has a single Arg residue between the mature sequence and a stop codon, as is the case with other teleost AM2/AM3 and AM5. These sequence characteristics support the classification of each eel AM by the phylogenetic analysis. Eel AM2 mRNA encoded a few types of mature peptides, which could be attributed to the repeated number of histidine (His) residue and base substitution (Fig. 2). As the NH2-terminal region did not affect the biological activity of human AM (33, 49), polymorphisms in this region of eel AM2 may not influence its function. In fact, we found that either type of AM2 polymorphism occurred equally in the cDNA sequences from nine eels.
Eel AM2 and AM5 induced profound hypotension when injected peripherally into eels, but AM(1) was much less potent than AM2 and AM5, as shown in this study. This is contrary to the results obtained in mammals in which vasodepressor effects of AM(1) are generally equipotent or even more potent than that of AM2 in rats (15, 16, 50, 63). Our preliminary data also showed that porcine AM5 is less potent than AM(1) for the vasodepressor effects when injected peripherally in rats. In eels, the decrease in PDA was profound, more than 50% from the preinjection level at 1.0 nmol/kg of AM2 and AM5. In mammals, however, AM(1) and AM2 decrease arterial blood pressure only by 10–20% and 20–30% at 1.0 and 10 nmol/kg, respectively (57, 63). Surprisingly, heterologous human AM2 has a potency and efficacy equivalent to eel AM2. Cardiovascular actions of mammalian ANP and ANG II were not as potent as that of homologous peptides in eels, despite high-sequence identities between eels and mammals (55, 58). In addition, pufferfish AM2 has a potency and efficacy similar to human AM2 for the vasodepressor effect in rats (8). It seems that the high potency of heterologous AM2 may be due to the slow evolutionary rate of AM2 as shown by the high-sequence identity between eel and human (81%). It is also possible that the evolution of the unidentified AM2 receptor is as slow as its ligand. On the other hand, bolus injection of heterologous human AM(1) did not change PDA at 0.01, 0.1, and 1.0 nmol/kg, probably because of the low sequence identity between eel and human (51%) (data not shown). The potent vasodepressor effects of AM2 and AM5 indicate its important role in cardiovascular regulation in teleost fish.
In fish, branchial and systemic circulation is connected in series, so that the effects on these sites can be differentiated by measuring changes in the pressure at the ventral aorta and the dorsal aorta (47). For instance, injection of AVT or oxytocin into the general circulation profoundly increases PVA but sometimes decreases PDA because they preferentially constrict branchial arterioles (28). Hypotension is primarily caused by the dilation of resistance vessels in the branchial and/or systemic circulation and by decreases in the stroke volume and heart rate (47). In mammals, an intravenous injection of AM preferentially decreases total peripheral resistance, resulting in a fall in arterial blood pressure (25, 32, 53). The hypotension is accompanied by increases in stroke volume and heart rate (25). Eel AMs decreased blood pressure more greatly in dorsal aorta than in ventral aorta, so that the hypotension seems to be attributable to a decrease in systemic resistance greater than in branchial resistance.
The decrease in arterial pressure was profound, ∼50% at 1.0 nmol/kg; however, the heart rate did not statistically change after injection of high doses of eel AM2 and AM5, except for those at a few points (Fig. 7). The vasodepressor effects of AM induce an increase in heart rate as observed in mammals (25). In trouts, urotensins I induced hypotension and increase heart rate by baroreflex when pretreated with prazosin (39). In fact, heart rate after injection of eel AMs indicated a tendency of increase. On the other hand, AMs may induce not only relaxation of blood vessels but also decrease of heart rate to decrease blood pressure. Although a direct action of AMs on heart rate is not well understood, it may compensate for the increase in heart rate induced by the baroreflex.
Eel AM2 and AM5 were more potent and efficacious than AM1 for the vasodepressor action, probably acting on the specific receptors other than AM1 receptor. In mammals, AM(1) binds CLR-RAMP2 and CLR-RAMP3 with high affinity and produces cAMP intracellularly after binding, whereas AM2 binds only CLR-RAMP3 with lower affinity than AM(1) (50, 57). This may account for the lower activity of AM2 in the peripheral actions in mammals (16, 50, 63). In teleost fish that have diversified CLRs and RAMPs, AM2 and AM5 bind CLR1-RAMP3, while AM1 binds not only CLR1-RAMP3 but also CLR1-RAMP2/5 and CLR2-RAMP2 with higher affinity (40). These findings are inconsistent with the more potent vasodepressor effects of AM2 and AM5 than AM1 in eels. Therefore, it is obvious that specific receptors for AM2 and AM5, which are different from the CLR-RAMP combination, should exist in the eel. In mammals, intracerebroventricular injection of AM and AM2 increases arterial blood pressure through sympathetic activation, and AM2 is more potent than AM, which is opposite to the peripheral actions (24, 63). Furthermore, as AM22–52, an antagonist of AM receptor, inhibited the central action of AM but only partially inhibited the AM2 action, the presence of specific receptor for AM2 has been suggested in the rat brain (24, 65). These findings suggest that the unidentified receptors for AM2 and AM5 should exist in vertebrates and participate in their peripheral hypotension in eels.
Eel AMs injected at 0.01 nmol/kg, near the minimal vasodepressor dose of eel AM2 and AM5, will increase plasma concentration by 6.5 × 10−11 M, if injected peptide is distributed quickly and evenly in the extracellular space of ∼155 ml/kg in eels (56). It seems that AMs injected into the circulation disappear slowly in eels, as the half-life of AM(1) in human plasma is rather long, ∼22 min (38), compared with ANP (1.5 min). Assuming that the concentration of AMs in eel plasma is comparable to those of AM(1) and AM2, 10−12 to 10−11 M, in the blood of human and rat (27, 63), the injected dose is close to the physiological range. In teleost fish, pufferfish AM2 and AM5 added to cultured cells expressing pufferfish CLR1 and RAMP3 increased cAMP production only slightly, even at 10−10 M (40). Therefore, circulating AM2 and AM5 may act on resistance vessels through unidentified receptors to maintain constant blood pressure in eels, although it is not known whether eel CLR and RAMP complex has similar affinities to homologous AMs as pufferfish. On the other hand, eel AMs injected at 1.0 nmol/kg, which results in plasma concentration to 6.5 × 10−9 M, may participate in the profound hypotension through binding to the CLR1 and RAMP3 complex also. Judging from the facts that AM2 and AM5 have profound vasodepressor effects in high concentration and ubiquitously express in various tissues, including vasculatures, it is also likely that they act on their receptors in a paracrine or autocrine fashion to induce vasorelaxation, although the vasodepressor effects through endocrine fashion cannot be excluded. In rats, AM2 receptors are localized in various vascular beds, as infusion of rat AM2 directly into the various organs decreased resistance (17).
To compare the potency and efficacy of AMs with other vasodepressor hormones, eel ANP, another potent hypotensive hormone in teleost fish (12, 48, 55), was injected in the current study. Eel AM2 and AM5 were almost equipotent with eel ANP in decreasing PDA, but they were much more efficacious than eel ANP and decreased PDA to less than half of the preinjection level at 1.0 nmol/kg. It has been suggested that eel ANP depresses arterial blood pressure through a decrease in venous return and branchial vasodilation rather than through peripheral vasodilation in eels, which contributes to the protection of the heart from volume overload (45, 48). Because excess decreases in venous return or branchial vasodilation will deteriorate cardiovascular and respiratory homeostasis, ANP may not induce as severe hypotension as do AM2 and AM5. On the other hand, as AM2 and AM5 may regulate blood pressure and flow at each organ by local vasodilation in a paracrine or autocrine fashion, which contrasts with the endocrine action of ANP, AM2, and AM5 may be able to depress PDA more profoundly than ANP.
In addition to the weak vasodepressor effects, eel AM1 at a high dose (5.0 nmol/kg) increased arterial pressure after an initial, transient hypotension in SW eels. The hypertension was not induced by eel AM2 and AM5 at the same high dose. It is shown that urotensins I (UI), a vasodilator in human and trout (26, 39), injected peripherally induced hypertension in eel and trout, which is shown to be mediated by catecholamine release from the chromaffin cells of the head kidney (7, 39). Eel and rat ANP also induces hypertension due to sympathetic activation in the trout (44), despite the fact that they are relaxed in isolated trout vessels (43, 54). Similar to the UI and ANP actions, the vasopressor effect of eel AM1 may be mediated by the adrenergic activation in the eel. In the rat, it is also suggested that peripherally injected AM(1) activates neurons in paraventricular nucleus (PVN) through the area postrema that has an incomplete blood-brain barrier (52). Injections of AM(1) and AM2 directly to the area postrema or the lateral ventricle increased arterial blood pressure in the rat (2, 63). It is possible that the AM-induced hypertension is mediated through the release of arginine vasopressin or oxytocin (24, 64) and activation of neurons in PVN (24). AVT, a homolog of arginine vasopressin of mammals, or oxytocin increases arterial blood pressure, particularly PVA, in the eel and the trout (11, 28, 48) through its specific action on the branchial vessels. However, as eel AM1 induced similarly effective hypertension in both PVA and PDA, the eel AM1-induced hypertension may not be mediated by AVT or isotocin, a homolog of oxytocin of mammals, secretion. Although it remained unknown whether AM1 evoked a hypertensive phase or not in FW eels, this result suggests that AM1 acts on cardiovascular system in eels through a signal cascade different from AM2/AM5.
To discuss the effects of salinity on the vasodepressor actions, regression lines of the decrease in PDA were calculated. The slope and intercept of regression lines were not significantly different between FW and SW eels. AMs seem to regulate blood pressure independently of habitat salinity. Eels make their habitat in FW, SW, and brackish water during their life cycle, which puts eels at risk of varying levels of body fluid conditions. In such a situation, AMs that are expressed in ubiquitous tissues may have important roles in the maintenance of blood pressure as well as neural control.
Tissue distribution of AM mRNAs expression was different between eels and pufferfish, T. rubripes (41). Ubiquitous expression of eel AMs may play important roles not only in maintenance of function in many tissues, but also in emergencies such as a large change of body fluid condition. In contrast, as pufferfish, T. rubripes, does not migrate within a wide range of salinity levels, they may not be exposed to large changes of body fluid condition. AM mRNAs of the pufferfish were expressed in selected tissues (41). In the pufferfish, AMs may be specialized to local regulation in the selected tissues. Although the vasodepressor effect of eel AM1 was not profound compared with other AMs, the strong expressions in heart and kidney are suggestive of the direct actions to cardiac myocyte and urinary excretion. Pufferfish AM2 and AM3 were expressed most abundantly in the brain (41), while eel AM2 and AM3 were weakly expressed in brain compared with other tissues. Central action of AM2 seems to increase arterial blood pressure in the rat (63). In the pufferfish, AM2 and AM3 may play an important role as a vasopressor hormone. Cardiovascular effects of AMs may have interspecific variability as that of ANP in teleost fish.
This study showed that AM2 and AM5 are potent vasodepressor hormones in eels. Takei et al. (61) suggested that vasodepressor hormones play more important roles than vasopressor hormones for the maintenance of low arterial pressure in aquatic fishes. This idea originates from the fact that the natriuretic peptide family is highly potent and much diversified in fishes (14, 31). The current study adds the AM family of peptides as such an example. Judging from the potent and efficacious vasodepressor effects of AMs in eels, endogenous AMs may have important roles in the maintenance of low arterial pressure together with other vasoactive peptides in teleost fish.
Perspectives and Significance
AM2 and AM5 were equally potent in the vasodepressor actions, although they have low sequence identity (39%) with each other. It is likely that they act on unidentified specific receptors to depress arterial pressure in eels as suggested in mammals. It is interesting to examine whether the two AMs share a common receptor or act on each specific receptor. This question may be clarified by the identification of novel receptors in eel vascular tissues. Since AM2 principally acts on the brain in mammals, the eel will be a good experimental animal for identification of the novel receptors for AM2 and AM5. In addition, the diversification of paralogous genes is of particular interest. The AM1, AM2, and AM5 genes were duplicated at the third round whole genome duplication that occurred around 335–440 million years ago during the teleost evolution (29). Among the duplicated genes, the AM4 gene duplicated from the AM1 gene has changed considerably, and the counterpart of duplicated AM5 gene has disappeared. However, the AM3 gene has been conserved extraordinarily compared with the AM2. AM3 may have similar vasodepressor effects and functions to AM2 in the eel. Therefore, it is possible that AM2 and AM3 share extremely important functions, or their genes are regulated by a different mechanism to differentiate their functions, which should be investigated in the future. There remain many issues to be addressed on the AM functions in eels, such as the central action, the mechanisms involved in the cardiovascular action, the renal effects, and the body fluid regulation of the AM family in fishes. To address each issue will give us deeper insights into the physiological significance of multiple AMs in teleost fishes.
This research was supported by Grant-in-Aid for Basic Research (A) from Japan Society for the Promotion of Science to Y. Takei (13304063 and 16207004).
We thank Dr. Susumu Hyodo, Dr. Koji Inoue, and Sanae Hasegawa of this laboratory for valuable comments and technical assistance.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society