Central and peripheral cardiovascular actions of adrenomedullin 5, a novel member of the calcitonin gene-related peptide family, in mammals
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
391
Central and peripheral cardiovascular actions of adrenomedullin 5,
a novel member of the calcitonin gene-related peptide
family, in mammals
Y Takei, H Hashimoto1, K Inoue, T Osaki2, K Yoshizawa-Kumagaye3, M Tsunemi3, T X Watanabe3,
M Ogoshi, N Minamino2 and Y Ueta1
Ocean Research Institute, University of Tokyo, Nakano, Tokyo 164-8639, Japan
1
Department of Physiology, School of Medicine, University of Occupational and Environmental Health, Kitakyushu, Fukuoka 807-8555, Japan
2
Department of Pharmacology, National Cardiovascular Center Research Institute, Suita, Osaka 565-8565, Japan
3
Peptide Institute Inc., Protein Research Foundation, Minoh, Osaka 562-8686, Japan
(Correspondence should be addressed to Y Takei; Email: takei@ori.u-tokyo.ac.jp)
Abstract
Adrenomedullin 5 (AM5) is a new member of the calcitonin approximately half as potent as AM. AM5 did not cause
gene-related peptide (CGRP) family identified in teleost fish. significant changes in urine flow and urine NaC concentration
Although its presence was suggested in the genome database of at any dose. In contrast to the peripheral vasodepressor action,
mammals, molecular identity and biological function of AM5 AM5 injected into the cerebral ventricle dose-dependently
have not been examined yet. In this study, we cloned a cDNA increased arterial pressure and heart rate at 0.1–1 nmol. The
encoding AM5 in the pig and examined its cardiovascular and increase reached maximum more quickly after AM5 (5 min)
renal effects. Putative mature AM5 was localized in the middle of than AM (15–20 min). AM5 added to the culture cells
prohormone and had potential signals for intermolecular ring expressing calcitonin receptor-like receptor (CLR) or calcitonin
formation and C-terminal amidation. The AM5 gene was receptor (CTR) together with one of the receptor activity-
expressed most abundantly in the spleen and thymus. Several modifying proteins (RAMPs), the combination of which forms
AM5 genes were newly identified in the database of mammals, major receptors for the CGRP family, did not induce
which revealed that the AM5 gene exists in primates, carnivores, appreciable increases in cAMP production in any combination,
and undulates but could not be identified in rodents. In although AM increased it at 10K10–10K9 M when added to the
primates, nucleotide deletion occurred in the mature AM5 CLR and RAMP2/3 combination. These data indicate that
sequence in anthropoids (human and chimp) during transition
AM5 seems to act on as yet unknown receptor(s) for AM5, other
from the rhesus monkey. Synthetic mature AM5 injected
than CLR/CTRCRAMP, to exert central and peripheral
intravenously into rats induced dose-dependent decreases in
cardiovascular actions in mammals.
arterial pressure at 0.1–1 nmol/kg without apparent changes in
Journal of Endocrinology (2008) 197, 391–400
heart rate. The decrease was maximal in 1 min and AM5 was
Introduction the third-round whole-genome duplication (3R) that
occurred in the teleost lineage (Vandepoele et al. 2004), but
Adrenomedullin (AM) has been known as a member of the the counterpart of AM5 may have disappeared during teleost
calcitonin gene-related peptide (CGRP) family that is evolution (Ogoshi et al. 2006). Since AM2/3 and AM5
composed of CGRP, AM, and amylin (López & Martı́nez should have existed when tetrapods (lobe-finned fishes)
2001, Brain & Grant 2004, Muff et al. 2004). However, AM diverged from ray-finned fishes, we searched them in the
was found to be diversified in teleost fish and forms an genome and expressed sequence tag (EST) databases and
independent subfamily consisting of five paralogs, AM1– identified AM2 and AM5 genes in mammals (Takei et al.
AM5 (Ogoshi et al. 2003). Synteny and phylogenetic analyses 2004b, Ogoshi et al. 2006). The AM2 gene was also identified
indicated that members of the AM subfamily can be divided in mammals by a similar comparative approach and named
into three groups, AM1/4, AM2/3, and AM5, and that intermedin (Roh et al. 2004, Takei 2006). In addition,
teleost AM1 is an ortholog of mammalian AM. Comparative another member of the CGRP family, calcitonin receptor-
genomic analyses further showed that two paralogs of each stimulating peptide (CRSP), was identified in the pig
group, AM1 and AM4, and AM2 and AM3, are generated at (Katafuchi et al. 2003), which was generated by tandem
Journal of Endocrinology (2008) 197, 391–400 DOI: 10.1677/JOE-07-0541
0022–0795/08/0197–391 q 2008 Society for Endocrinology Printed in Great Britain Online version via http://www.endocrinology-journals.org
Downloaded from Bioscientifica.com at 11/11/2021 01:24:26PM
via free access392 Y TAKEI and others . Cardiovascular action of adrenomedullin 5
duplication of the CGRP gene (Ogoshi et al. 2006). Thus, the species (Allen et al. 1997, Taylor et al. 2005, Hashimoto et al.
CGRP family appears to be more diversified than it was 2007). Finally, specific AM5 receptors were sought in culture
previously thought. cells expressing homologous porcine CTR/CLR and RAMP
All CGRP family peptides exhibit cardiovascular actions, using cAMP production as a marker.
although relative potency differs among the members,
CGRPOAMRAM2[amylin (Ando et al. 1990, Charles
et al. 1997, Hall & Brain 1999, Fujisawa et al. 2004). CGRP is
a neuropeptide in the brain and periphery, which is released Materials and Methods
from axon terminals innervating the vascular smooth muscles
(see Brain & Grant 2004 for review). AM is synthesized in Molecular studies
both vascular endothelial cells and smooth muscle cells and
exerts potent hypotensive actions by relaxation of microvessels All animal experiments reported in this paper have been
in various peripheral tissues (see López & Martı́nez (2001) for approved by the Committee for Animal Experiments of the
review). Expression of the AM gene is enhanced in various University of Tokyo and of the University of Occupational
forms of cardiac failure and renal dysfunction, causing and Environmental Health. Immature female pig of 7.8 kg
protective actions on the heart and kidney (Tsuruda & was purchased from a local dealer. After decapitation, the
Burnett 2002). The AM2 gene is expressed in the brain and brain, pituitary, heart, lung, thymus, spleen, kidney, adrenal,
kidney of mammals and exerts various central and peripheral stomach, and liver were excised, cut into small pieces, and
actions on cardiovascular and body fluid regulation (Roh et al. frozen in liquid nitrogen. Total RNA was extracted from the
2004, Takei et al. 2004a, Taylor et al. 2005, Yang et al. 2005, frozen tissues using Isogen (Nippon Gene, Toyama, Japan).
Takahashi et al. 2006). Amylin, a pancreatic hormone that is A double-stranded cDNA pool was prepared from 1 mg total
secreted with insulin from b cells (Cluck et al. 2005), exhibits RNA from the spleen using SMART cDNA library
a weak vasorelaxant effect (1/100 of CGRP). construction kit (Clontech). The whole coding region of
Biological actions of the CGRP family peptides are pig AM5 cDNA was obtained by PCR, under the condition
principally mediated by the complex of calcitonin receptor described previously (Ogoshi et al. 2003), using primers that
(CTR) or CTR-like receptor (CLR) associated with one of the correspond to putative 5 0 - and 3 0 -untranslated region
three receptor activity-modifying proteins (RAMPs); CLR– (Table 1). Amplified fragments were subcloned and
RAMP1 is a receptor for CGRP, CLR–RAMP2/3 for AM, sequenced more than ten clones.
CLR–RAMP3 for AM2, and CTR–RAMP2 for amylin (Brain The tissue distribution of AM5 transcripts was examined by
& Grant 2004, Conner et al. 2004). After ligand binding, the RT-PCR. Total RNA from each tissue (2 mg) was reverse
receptor complex increases intracellular cAMP to mediate transcribed as above, and PCR was performed using gene-
biological actions. However, AM2 has lower affinity than AM to specific primers for porcine AM5 and glyceraldehyde-3-
the CLR–RAMP3 complex, while central action of AM2 was
phosphate dehydrogenase (GAPDH) that was used for
more potent than AM in the rat (Hashimoto et al. 2007).
internal control (Table 1). Annealing temperature and
Furthermore, the central AM2 effect was not blocked by
number of amplification cycles were 65 8C and 40 cycles
AM22–52 and CGRP8–37, antagonists for CGRP family
for AM5 and 58 8C and 35 cycles for GAPDH.
receptors. Therefore, a specific receptor for AM2 other than
CLR–RAMP3 may exist in the rat (Taylor et al. 2006). AM5 genes were sought in the genome and EST databases
Judging from the potent cardiovascular actions of AM and of various vertebrate species using BioGrepX program
AM2 in mammals, a new member of the AM subfamily, established by Dr Hideo Bannai of Kyushu University (see
AM5, may also have similar actions through the CLR/CTR– Takei et al. 2004b). Phylogenetic analyses of newly identified
RAMP complex. However, only the presence of the AM5 AMs were performed using a Bayesian method in MrBayes
gene is predicted in the database and it is not known whether program (version 3.1.2; Ronquist & Huelsenbeck 2003) to
it is actually expressed as a functional protein in mammalian confirm their identity in the AM subfamily.
tissues. In this study, we cloned a cDNA encoding AM5 and
examined the tissue distribution of transcripts in the pig.
Then, inferred mature AM5 was chemically synthesized and Table 1 List of primers used for cloning and RT-PCR analyses
its cardiovascular effects examined after peripheral and central
Sequence
injections in the rats. In parallel with the cardiovascular
actions, we examined the renal effect of AM5 as AM and Name
AM2 have diuretic and natriuretic actions in the rats (Fujisawa GAPDH-A4 5 0 -CAGCATCAAAGGTAGAAGAGTGAGTGT-3 0
et al. 2004). We used rats as an experimental species, because GAPDH-S3 5 0 -TACATGGTCTACATGTTCCAGTATGA-3 0
we have an established technique to examine cardiovascular pigAM5-A1 5 0 -GAGTTCTCGTATATTTATTCAGCTCTGGA-3 0
pigAM5-A2 5 0 -GGAGTTAGGGCAAAAGAAACTTGAGGT-3 0
and renal effects in rats (Watanabe et al. 1988) and because pigAM5-S1 5 0 -ACACCCGTGAGGGCTCCGTA-3 0
most cardiovascular effects of AM and AM2 after peripheral pigAM5-S2 5 0 -GTGGCTCTTTCGGCCTCAGT-3 0
and central administrations have been investigated in this
Journal of Endocrinology (2008) 197, 391–400 www.endocrinology-journals.org
Downloaded from Bioscientifica.com at 11/11/2021 01:24:26PM
via free accessCardiovascular action of adrenomedullin 5 . Y TAKEI and others 393
Physiological studies Porcine CTR, CLR, and three isoforms of RAMP cDNAs
encoding complete open reading frames were isolated from
Predicted mature peptide of porcine AM5, which consists of
the porcine lung and hypothalamus cDNA libraries and
50 amino acid residues with an intramolecular ring formed by
ligated into pcDNA 3.1 (C) expression vector as described
a disulfide bond and an amidated C terminus, was synthesized
previously (Katafuchi et al. 2003). The CLR or CTR either
by a peptide synthesizer (Applied Biosystems, 430A) with
alone or in combination of one of the three RAMPs (cDNA
p-methyl-benzhydrylamine resin as a solid support. The ratio 1:4) were co-transfected into COS-7 cells with
correct sequence was confirmed by mass analysis, amino acid Lipofectamine Plus reagent (Invitrogen Corp.) according to
analysis, and reverse-phase HPLC. Human AM was purchased the manufacturer’s protocol. The transfected cells were then
from the Peptide Institute Inc. (Osaka, Japan). We used used for cAMP assay.
human AM to compare the effect with porcine AM5 because COS-7 cells were plated at 40 000 cells/well in 48-well
porcine AM was not commercially available and human plate and cultured for 24 h. Subsequently, the transfected cells
and porcine AM differs by only one of 52 amino acid residues. were washed twice with Dulbecco’s modified Eagle’s medium
For i.v. injection experiments, adult male Sprague–Dawley dissolved in 20 mM HEPES, pH 7.4, containing 0.5 mM
rats weighing 285.3G3.9 g (nZ7) were purchased from a 3-isobutyl-1-methylxanthine and 0.05% BSA, followed by
commercial source. After anesthesia, cannulae were inserted in incubation in this medium for 30 min at 37 8C. The
the femoral artery, femoral vein, and urinary bladder (Watanabe incubation medium was then replaced by 200 ml medium
et al. 1988). The arterial cannula was connected to a pressure containing 10K11–10K6 M of hormones, and incubated at
transducer and tachometer for continuous measurement of 37 8C for another 30 min. Aliquots (100 ml) of the medium
arterial pressure and heart rate. The venous cannula was were used for cAMP measurement by RIA as reported
connected to a syringe for continuous infusion of a Ringer previously (Katafuchi et al. 2003).
solution (NaCl, 130; KCl, 5; CaCl2, 5.3; NaHCO3, 2 in mM)
and for injection of peptides. Urine was collected every 10 min,
and its volume and NaC and ClK concentrations were Statistical analyses
determined. AM5 and AM were injected as a bolus at 0.1, All data were expressed as meansGS.E.M. Time-course
0.3, and 1 nmol/kg in 0.5 ml NaCl solution containing 0.01% changes in cardiovascular and renal parameters after injection
Triton X-305 (nZ7). Vehicle served as control. Injection of AM5 were statistically compared with controls by the
interval was 20 min at 0.1 and 0.3 nmol/kg, and 40 min at ANOVA, followed by the Tukey’s test at each time point. The
1 nmol/kg. Arterial pressure returned to a pre-injection level in dose–response relationship was examined by the Dunnett’s
10 min after injection of the highest dose. test. In case normal distribution was not demonstrable, Steel–
For i.c.v. injection experiments, adult male Wistar rats Dwass test was used for comparison (Takagi et al. 2003).
(273.8G9.8 g, nZ34) were anesthetized, and stainless steel Significance was set at P!0.05.
cannula implanted stereotaxically (o.d.Z0.55 mm) aimed at the
left lateral ventricle at the following coordinates: 0.8 mm
posterior to the bregma, 1.4 mm lateral to the midline, and
2.0 mm below the surface of left cortex such that a tip of the Results
cannula was 1.0 mm above the left cerebral ventricle
(Hashimoto et al. 2005). Two anchoring screws were fixed to Molecular studies
the skull and the cannula was secured in place by acrylic dental A cDNA coding for porcine AM5 was cloned from the spleen
cement. Seven days postoperation, the animals were anesthe- (Fig. 1A). The cDNA had a single nucleotide polymorphism
tized with urethane (1.4 g/kg) and catheter inserted into the (C and T) in the coding region, which changes amino acid
femoral artery for continuous recording of arterial pressure. The from Ser to Phe in the mature sequence. As both types of
arterial pressure and heart rate were recorded before and after cDNAs occurred in equal numbers, they may originate from
i.c.v. injection of AM5 or AM at doses of 0, 0.1, 0.3, or 1 nmol/ two haploid genomes of parents. Prohormone excluding a
rat in 10 ml. After experiment, i.c.v. injection was confirmed by putative signal peptide consists of only 91 amino acid residues,
a dye injection. and mature AM5 may be cleaved off by furin at a typical
processing signal (Arg-X-X-Arg) in the N-terminal region
Cyclic AMP accumulation in culture cells expressing CGRP and at consecutive Arg residues that connect the C-terminal
peptide (Fig. 1A). Accordingly, mature AM5 of 50 amino acid
receptors
residues may be formed after disulfide bond formation by two
Synthetic porcine calcitonin was purchased from Bachem AG Cys residues in the N-terminal region and C-terminal
(Bubendorf, Switzerland), and human AM from Peptide amidation using a Gly residue after removal of C-terminal
Institute Inc. Porcine CGRP was custom synthesized by Arg residues. The AM5 gene was expressed abundantly in the
American Peptide Company Inc. (Sunnyvale, CA, USA). The spleen and thymus, and slightly in the adrenal and pituitary
sequence identity of porcine and human AM is 94%, while those (Fig. 1B). The signal was undetectable in the brain, heart,
of CGRP and calcitonin are 84 and 44% respectively. kidney, liver, and stomach.
www.endocrinology-journals.org Journal of Endocrinology (2008) 197, 391–400
Downloaded from Bioscientifica.com at 11/11/2021 01:24:26PM
via free access394 Y TAKEI and others . Cardiovascular action of adrenomedullin 5
Figure 1 (A) Nucleotide and predicted amino acid sequences of pig adrenomedullin 5 (AM5). Putative
signal sequence is underlined. Arrows indicate the processing site for production of mature peptide. Bold
letters with underline show the site of polymorphism. (B) Expression of the AM5 gene in porcine tissues.
GAPDH transcript was used as internal control.
We identified AM5 sequences in the genome database of are more conserved in mammals than in teleost fish (Fig. 2).
ungulates (artiodachtyls; pig, ox, and sheep and perissodactyles; Xenopus AM5 is more similar to teleost AM5 than to
horse), carnivores (dog and cat), primates (prosimians; Tupaia mammalian AM5.
belangeri and simians; Macaca malatta) (Fig. 2), but we could not
identify such sequences in rodents (mice and rats). In primates,
Physiological studies
AM5-like sequence exists in human (AC011495) and in the
chimp (Pan troglodytes, NW_001228245), but the sequence Cardiovascular and/or renal parameters in rats before AM
changes in the middle of mature peptide because of the deletion injections are shown in Table 2. While i.v. injections of AM5
of two nucleotides. The AM5 sequences were highly conserved and AM decreased arterial pressure immediately, i.c.v. injections
in mammals. In amphibians, AM5 was identified in two Xenopus increased it slowly and for a longer period (Fig. 3A and B).
species, X. tropicalis and X. laevis (Fig. 2). The putative mature Control vehicle injection did not alter arterial pressure in both
sequences differed from each other by four amino acid residues. i.v. and i.c.v. injections. The peak increase occurred 5 min after
In teleost fish, AM5 sequences were identified in nine i.c.v. injection of AM5, and the increase was slower and lasted
species (Fig. 2). The AM5 sequence is fairly variable (83% longer after AM injection. Even with such a profound
identity) in two species of pufferfish (Takifugu rubripes and hypotension after i.v. injection of AM5, heart rate did not
Tetraodon nigroviridis), but it is identical in two salmonid fishes increase simultaneously at the highest dose (Fig. 3C). After i.c.v.
(Oncorhyncus mykiss and Salmo salar). In general, AM5 sequences injection, on the other hand, heart rate increased profoundly,
Journal of Endocrinology (2008) 197, 391–400 www.endocrinology-journals.org
Downloaded from Bioscientifica.com at 11/11/2021 01:24:26PM
via free accessCardiovascular action of adrenomedullin 5 . Y TAKEI and others 395
Figure 2 Amino acid sequences of adrenomedullin 5 thus far identified in vertebrates. Bracket shows
a disulfide bond. Amino acid residues identical in more than half the species are shaded. Accession numbers:
pig, AB287333; ox, AV664186; sheep, EE827578; horse, NW_001800322; dog, DN365983; cat,
AANG01678840; tupai (Tupaia belangeri ), scaffold_50755; monkey (Macaca mulatta), NW_001106523;
Xenopus laevis, CD301659; Xenopus tropicalis, DN011652; Takifugu rubripes, AB120299; Tetraodon
nigroviridis, SCAF14992; flounder (Paralichthys olivaceus), AU091250; medaka, AB257078; zebrafish,
NW_634155; rainbow trout, BX878389; Atlantic salmon, DY716166; eel (Anguilla japonica), AB363988.
and the peak occurred more quickly with AM5 than with AM as cAMP production with CLR–RAMP1, but the increase was
is the case for the increase in arterial pressure (Fig. 3D). significant only at 10K8 M for AM and 10K7 M for AM5.
The dose–response analyses showed that AM5 was approxi- AM increased cAMP production with CLR–RAMP2 at
mately half as potent as AM in terms of peripheral vasodepressor 10K10 M and CLR–RAMP3 at 10K9 M, but AM5 was much
effect (Fig. 4A). The increase in heart rate was significant at less potent and efficacious than AM (Fig. 6C and D). AM5
1 nmol/kg AM, probably because of the reflex tachycardia in and AM increased cAMP accumulation slightly in cells
response to hypotension (Fig. 4B). The urine volume tends to co-expressing CTR and RAMP, with AM more potent and
decrease at 1 nmol/kg, which may be due also to the profound efficacious than AM5 (Fig. 7). Calcitonin increased cAMP
hypotension, while urinary NaC excretion did not change after accumulation in cells co-expressing CTR and/or RAMP at
i.v. injections of AM5 and AM (Fig. 4C and D). Urinary ClK 10K10 M and the increase was more than tenfold at 10K7 M
excretion did not change either (data not shown). The dose– (Fig. 7).
response analyses for the i.c.v. vasopressor effect showed that
AM5 is more potent than AM 5 min after injection (Fig. 5A), Table 2 Cardiovascular and renal parameters in Nembutal-
but the relationship was reversed at 20 min because the pressure anesthetized rats infused with Ringer and used for i.v. injection
(nZ7) and cardiovascular parameters in urethane-anesthetized rats
returned to the pre-injection level 20 min after AM5 injection used for i.c.v. injection (nZ34)
(Fig. 5B). When the maximal increase was compared, the
vasopressor potency was comparable between AM5 and AM. i.v. injection i.c.v. injection
The tendency was similar with chronotropic effect; AM5 was
Parameters
more effective at 5 min and AM was more effective at 20 min Arterial pressure (mmHg) 112.4G8.7 74.9G9.9
(Fig. 5C and D). Heart rate (beat/min) 428.3G22.1 350.8G22.8
Neither AM5, AM, nor CGRP increased cAMP concen- Urine volume (ml/h) 400.1G48.9 –
tration appreciably in the medium when added to the culture Urine NaC concentration (mM) 46.1G8.6 –
Urine ClK concentration (mM) 70.1G13.8 –
cells that express porcine CLR alone (Fig. 6A), but CGRP Urinary NaC excretion (mmol/h) 15.3G3.6 –
increased it profoundly in cells co-expressing CLR and Urinary ClK excretion (mmol/h) 24.6G6.1 –
RAMP1 at 10K11 M (Fig. 6B). AM and AM5 also increased
www.endocrinology-journals.org Journal of Endocrinology (2008) 197, 391–400
Downloaded from Bioscientifica.com at 11/11/2021 01:24:26PM
via free access396 Y TAKEI and others . Cardiovascular action of adrenomedullin 5
Figure 3 Time-course changes in (A and B) arterial pressure and (C and D) heart rate after i.v. injection of
0.3 nmol/kg and i.c.v. injection of 1 nmol/rat of adrenomedullin 5 (open circles), adrenomedullin (open
squares), or vehicle (closed circles) in the rat. *P!0.05 compared with the value of controls.
Discussion Calcitonin, hypocalcemic hormone produced by alternative
splicing of the CGRP gene, is also occasionally included in
The CGRP family peptides have been shown to exert wide the CGRP family (Muff et al. 2004).
spectra of biological actions in various tissues and have The evolutionary history of the CGRP family has been
attracted attention of both basic and clinical endocrinologists unveiled in fishes and tetrapods by comparative genomic analyses
because of their important roles in various forms of diseases. (Ogoshi et al. 2006). It has been suggested that CGRP–AM(1),
For example, CGRP seems to be a major cause of migraine amylin–AM2, and AM5 existed on three different proto-
through local vasodilatation of cerebral vessels by release from chromosomes before divergence of ray-finned fishes and lobe-
trigeminal sensory nerve terminals (Arulumani et al. 2004). finned fishes that lead to tetrapods. In teleosts, all genes were
The AM gene is up-regulated in response to cardiac and renal duplicated at the 3R, and thus AM(1) and AM4, AM2 and AM3,
failure or septic shock in proportion to the severity of the and CGRP1 and CGRP2 genes were produced in the teleost
disease state (Eto et al. 2003, Rademaker et al. 2003). Amylin, lineage. One of the duplicated amylin and AM5 genes appear to
co-secreted with insulin from pancreatic b cells, is now used have been silenced after the 3R. Before the 3R, two
for the treatment of type 2 diabetes (Cluck et al. 2005). The chromosomes on which CGRP–AM(1) and amylin–AM2
second AM, named AM2/intermedin, was recently identified exist may be duplicated between CGRP and amylin and
in the selected species of mammals based on the discovery of between AM(1) and AM2 at the second-round whole-genome
AM subfamily in teleost fish (Roh et al. 2004, Takei et al. duplication that occurred during transition from agnathans to
2004b). AM2 seems to be a multifunctional peptide as is AM gnathostomes. The origin of the AM5 gene is not known, but it
but has more potent central actions than AM (Taylor et al. seems to be produced from the AM(1) or AM2 gene as judged by
2005, Hashimoto et al. 2007). CRSP is a brain peptide that the high sequence similarity between AM5 and AM(1)/AM2.
stimulates CTR alone, and its function is now under AM5 sequences were highly conserved within mammalian
investigation (Katafuchi et al. 2003). In addition, the present species. However, the homology is low between mammals
study showed that the AM5 gene exists in primates, and teleost fishes, although CGRP, AM(1), AM2, and amylin
carnivores, and ungulates and is expressed as mRNA in the sequences are well conserved even across different classes
pig tissues. Furthermore, AM5 exhibited brisk cardiovascular of vertebrate. Xenopus AM5 sequences were more similar to
actions when administered in the brain or periphery as did teleost AM5 than to mammalian AM5, although amphibians
AM and AM2. Therefore, the CGRP family is now consisted are phylogenetically closer to mammals than to teleost
of CGRP, AM, amylin, AM2, CRSP, and AM5 in mammals. and have orthologous hormones similar to mammals
Journal of Endocrinology (2008) 197, 391–400 www.endocrinology-journals.org
Downloaded from Bioscientifica.com at 11/11/2021 01:24:26PM
via free accessCardiovascular action of adrenomedullin 5 . Y TAKEI and others 397
Figure 4 Dose–response relationship for the effects of adreno- Figure 5 Dose–response relationship for the effects of adreno-
medullin 5 (open circles) and adrenomedullin (closed circles) on medullin 5 (open circles) and adrenomedullin (closed circles) on
(A) arterial pressure, (B) heart rate, (C) urine volume, and (D) urinary (A and B) arterial pressure and (C and D) heart rate 5 min (A and C)
NaC excretion after i.v. injection in the rat. *P!0.05 compared and 20 min (B and D) after i.c.v. injection in the rat. *P!0.05
with the value of vehicle injection. compared with the value of vehicle injection.
www.endocrinology-journals.org Journal of Endocrinology (2008) 197, 391–400
Downloaded from Bioscientifica.com at 11/11/2021 01:24:26PM
via free access398 Y TAKEI and others . Cardiovascular action of adrenomedullin 5
Figure 6 Cyclic AMP accumulation in the medium of COS-7 cells expressing (A) calcitonin receptor-like
receptor (CLR) alone, (B) CLRCreceptor activity-modifying protein (RAMP)1, (C) CLRCRAMP2, and
(D) CLRCRAMP3 after administration of adrenomedullin 5 (open circles), adrenomedullin (closed
circles), or calcitonin gene-related peptide (open squares). Each point represents the meanGS.E.M. of
three separate determinations. *P!0.05 compared with control.
(e.g. Sherwood et al. 2000). It seems that Xenopus AM5 has in vitro was not demonstrated by AM (Taylor et al. 2006).
specific sequences that are important for the aquatic life. In Therefore, the presence of AM2-specific receptor(s) that differs
primates, nucleotide deletion occurred in the AM5 gene in from CLR–RAMP3 has been suggested. In this study, we
the region that codes for mature sequence during the showed that AM5 is much less potent and efficacious than AM in
transition from rhesus monkey (Macaca) to anthropoids any CTR/CLR and RAMP combinations, although cardio-
(chimp and human). In anthropoids, the AM5 gene with vascular effect of AM5 after i.v. and i.c.v. injections were
deletion seems to be expressed and registered as carnitine comparable with that of AM. Thus, AM5 seems to have yet
acyltransferase-like protein 1 in the EST database unidentified specific receptor(s) that differ from CTR and CLR.
(AF331918). The AM5 gene was not detectable in mice A comparative study also suggests the presence of specific
and rats, but the potent cardiovascular actions shown in this receptors for AM2 and AM5. In the eel, homologous eel
study indicate that the gene is present with sequences altered AM2 and AM5 are 100-fold more potent and efficacious than
at conserved amino acid residues in rodents, or it was silenced AM(1) for the vasodepressor effect when injected into the
recently in rodents as in anthropoids. circulation (Nobata et al. 2008). In the pufferfish (Takifugu
It is suggested that the CGRP family peptides act principally obscurus), three CLRs and five RAMPs have been identified
in a paracrine fashion to exhibit biological activity (Brain & (Nag et al. 2006). However, homologous AM5 and AM2
Grant 2004). Judging from the high affinity of CGRP, AM, and binds only to CLR1–RAMP3 with low affinity, while AM(1)
amylin to CTR/CLR coupled with one of RAMPs (10K11– bind to CLR1–RAMP2/3/5 and CLR2–RAMP2 com-
10K9 M), which is in a range of endogenous variation of plasma binations with much higher affinity as determined by the
concentrations, these peptides may act also as an endocrine COS cells expressing these proteins. Therefore, specific
hormone from plasma (Eto et al. 2003). However, the potency of receptors for AM5 and AM2 should exist in the vascular
AM2 for cAMP accumulation in CLR–RAMP was much smooth muscles of teleosts. It seems that teleost fish serve as
lower than AM (Roh et al. 2004, Takei et al. 2004a), while its excellent materials for identification of new AM5 and/or
central actions are even greater than AM (Hashimoto et al. AM2 receptor other than CLR–RAMP in vertebrates.
2007). In addition, the AM2 effect was blocked only partially by AM5 caused immediate hypotension when injected into
CGRP8–37 and AM22–52, although the AM effect was totally the periphery and caused long-lasting hypertension when
blocked by combination of the two blockers. Furthermore, the injected centrally in this study as observed with AM. The
inhibitory effect of AM2 on GH secretion from pituitary cells hypertension and tachycardia after i.c.v. injection of AM5
Journal of Endocrinology (2008) 197, 391–400 www.endocrinology-journals.org
Downloaded from Bioscientifica.com at 11/11/2021 01:24:26PM
via free accessCardiovascular action of adrenomedullin 5 . Y TAKEI and others 399
Figure 7 Cyclic AMP accumulation in the medium of COS-7 cells expressing (A) calcitonin receptor
(CTR) alone, (B) CTRCreceptor activity-modifying protein (RAMP)1, (C) CTRCRAMP2, and (D) CTRC
RAMP3 after administration of adrenomedullin 5 (open circles), adrenomedullin (closed circles), or
calcitonin (open squares). Each point represents the meanGS.E.M. of three separate determinations.
*P!0.05 compared with control.
may be due to the sympathetic activation as shown by AM AM5 transcripts were identified in the spleen, head kidney
and AM2 (Taylor et al. 2005, Hashimoto et al. 2007). The (hematopoietic tissue equivalent to bone marrow), gill, and
peripheral vasodepressor effect of AM5 is half as potent as skin, all of which are implicated in defensive functions
AM, but the central vasopressor effect was comparable against infection (Ogoshi et al. 2003). Therefore, AM5 is
between the two peptides. It is possible that the peripheral expressed abundantly in the defense-related organs or
vasodepressor effect was ameliorated by the central vasopres- hematopoietic organs in both mammals and fishes. The AM
sor action mediated via the circumventricular organ that has family of peptides have diverse functions that include
incomplete blood–brain barrier. It has been shown that AM regulation of cardiovascular, body fluid, and immune systems
acts on the area postrema, one of such organs, to elevate (López & Martı́nez 2000, Brogden et al. 2005), of which
arterial pressure (Allen et al. 1997), and the area postrema AM(1) is potent in peripheral actions and AM2 in central
possesses AM receptors and responds to AM by modulating actions. It is intriguing to examine what is the major function
excitability of the neurons (Yang & Ferguson 2003). The of AM5 in the AM family. The AM5 gene was disrupted very
difference in the time course of central vasopressor effect also recently in human and may be silenced in rodents. Therefore,
indicates that AM5 and AM may act on different receptors or it is of interest to examine how the loss of the AM5 gene
use different signal transduction systems. The lack of the renal has influenced their biological systems, particularly in the
effect may be due to the injection of hormones into the immune and hematopoietic system.
general circulation as AM was diuretic and natriuretic only
when infused directly into the renal artery in rats (Fujisawa
et al. 2004). The concomitant hypotension after i.v. injection Acknowledgements
of AM5 may have decreased GFR and thus ameliorated the
possible diuretic effect. The renal effect of AM5 may be minor We thank Dr Junyou Li of Animal Resource Science Center,
compared with ANP because ANP is diuretic and natriuretic Graduate School of Agricultural and Life Sciences, the
after a bolus systemic injection or infusion even with University of Tokyo and Dr Tadaomi Naka of Ocean
concomitant hypotension (Watanabe et al. 1988). Research Institute for their help in sampling pig tissues. We
The present study showed that the AM5 gene is expressed also thank Ms. Sanae Hasegawa for measurements of urinary
abundantly in the spleen and thymus of pig. In the pufferfish, ion concentrations. This work was supported in part by a
www.endocrinology-journals.org Journal of Endocrinology (2008) 197, 391–400
Downloaded from Bioscientifica.com at 11/11/2021 01:24:26PM
via free access400 Y TAKEI and others . Cardiovascular action of adrenomedullin 5
grant from the Japan Society for the Promotion of Science Nobata S, Ogoshi M & Takei Y 2008 Potent cardiovascular actions of
(16207004) to Y T. There is no conflict of interest that would homologous adrenomedullins in eel. American Journal of Physiology (In press).
Ogoshi M, Inoue K & Takei Y 2003 Identification of a novel adrenomedullin
prejudice its impartiality. gene family in teleost fish. Biochemical and Biophysical Research Communications
311 1072–1077.
Ogoshi M, Inoue K, Naruse K & Takei Y 2006 Evolutionary history of the
References calcitonin gene-related peptide family in vertebrates revealed by
comparative genomic analyses. Peptides 27 3154–3164.
Allen MA, Smith PM & Ferguson AV 1997 Adrenomedullin microinjection Rademaker MT, Cameron VA, Charles CJ, Lainchbury JG, Nicholis MG &
into the area postrema increases blood pressure. American Journal of Richards AM 2003 Adrenomedullin and heart failure. Regulatory Peptides
Physiology 272 R1698–R1703.
112 51–60.
Ando K, Pegram BL & Frohlich ED 1990 Hemodynamic effects of calcitonin
Roh J, Chang CL, Bhalla A, Klein C & Hsu SY 2004 Intermedin is a
gene-related peptide in spontaneously hypertensive rats. American Journal of
calcitonin/calcitonin gene-related peptide family peptide acting through
Physiology 258 R425–R429.
the calcitonin receptor-like receptor/receptor activity-modifying protein
Arulumani U, Maassenvandenbrink A, Villalon CM & Saxena PR 2004
receptor complexes. Journal of Biological Chemistry 279 7264–7274.
Calcitonin gene-related peptide and its role in migraine pathophysiology.
Ronquist F & Huelsenbeck JP 2003 MrBayes 3: Bayesian phylogenetic
European Journal of Pharmacology 500 315–330.
inference under mixed models. Bioinformatics 19 1572–1574.
Brain SD & Grant AD 2004 Vascular actions of calcitonin gene-related peptide
and adrenomedullin. Physiological Reviews 84 903–934. Sherwood NM, Krueckl SL & McRory JE 2000 The origin and function of
Brogden KA, Guthmille JM, Salzet M & Zasloff M 2005 The nervous system the pituitary adenylate cyclase-activating polypeptide (PACAP)/glucagon
and innate immunity: the neuropeptide connection. Nature Immunology 6 superfamily. Endocrine Reviews 21 619–670.
558–564. Takagi T, Okamoto K, Yokogawa Y, Yokota M, Kurokawa K & Yasunaga T
Charles CJ, Rademaker MT, Richards AM, Cooper GJ, Coy DH, Jing NY & 2003 A new multiple comparison method using resampling technique for
Nicholls MG 1997 Hemodynamic, hormonal, and renal effects of medical, pharmaceutical, and chemical data. Journal of Computer Aided
adrenomedullin in conscious sheep. American Journal of Physiology 272 Chemistry 4 27–34.
R2040–R2047. Takahashi K, Kikuchi K, Maruyama Y, Urabe T, Nakajima K, Sasano H, Imai
Cluck MW, Chan CY & Adrian TE 2005 The regulation of amylin and Y, Murakami O & Totsune K 2006 Immunocytochemical localization of
insulin gene expression and secretion. Pancreas 30 1–14. adrenomedullin 2/intermedin-like immunoreactivity in human hypo-
Conner AC, Simms J, Hay DL, Mahmoud K, Howitt SG, Wheatley H & thalamus, heart and kidney. Peptides 27 1383–1389.
Poyner DR 2004 Heterodimers and family-B GPCRs: RAMPs, CGRP Takei Y 2006 Adrenomedullin 2/intermedin. In Handbook of Biologically Active
and adrenomedullin. Biochemical Society Transactions 32 843–846. Peptides, pp 1281–1286. Ed. AJ Kastin. Amsterdam: Elsevier Press.
Eto T, Kato J & Kitamura K 2003 Regulation of production and secretion of Takei Y, Hyodo S, Katafuchi T & Minamino N 2004a Novel fish-derived
adrenomedullin in the cardiovascular system. Regulatory Peptides 112 61–69. adrenomedullin in mammals: Structure and possible function. Peptides 25
Fujisawa Y, Nagai Y, Miyatake A, Takei Y, Miura K, Shoukouji T, Nishiyama 1643–1655.
A, Kimura S & Abe Y 2004 Renal effects of a new member of Takei Y, Inoue K, Ogoshi M, Kawahara T, Bannai H & Miyano S 2004b
adrenomedullin family, adrenomedullin 2, in rats. European Journal of Identification of novel adrenomedullin in mammals: a potent cardiovascular
Pharmacology 497 75–80. and renal regulator. FEBS Letters 556 53–58.
Hall JM & Brain SD 1999 Interaction of amylin with calcitonin gene-related Taylor MM, Bagley SL & Samson WK 2005 Intermedin/adrenomedullin-2
peptide receptors in the microvascularture of the hamster cheek pouch acts within central nervous system to elevate blood pressure and inhibit food
in vivo. British Journal of Pharmacology 126 280–284. and water intake. American Journal of Physiology 288 R919–R925.
Hashimoto H, Hyodo S, Kawasaki M, Mera T, Soya A, Saito T, Fujihara H, Taylor MM, Bagley SL & Samson WK 2006 Intermedin/adrenomedullin-2
Takei Y & Ueta Y 2005 Centrally administered adrenomedullin 2 activates inhibits growth hormone release from cultured, primary anterior pituitary
hypothalamic oxytocin – secreting neurons causing elevated plasma cells. Endocrinology 147 85–964.
oxytocin level in rats. American Journal of Physiology 289 E753–E761. Tsuruda T & Burnett JC 2002 Adrenomedullin – an autocrine/paracrine
Hashimoto H, Hyodo S, Kawasaki M, Shibata M, Saito T, Suzuki H, Otsubo H, factor for cardiorenal protection. Circulation Research 90 625–627.
Yokoyama T, Fujihara H, Higuchi Tet al. 2007 Adrennomedullin 2 is a more Vandepoele K, De Vos W, Taylor J, Meyer A & Van de Peer Y 2004 Major
potent activator of hypothalamic oxytocin-secreting neurons than adreno- events in the genome evolution of vertebrates: paranome age and size differ
medullin in rats, and its effects are only partially blocked by antagonists for considerably between ray-finned fishes and land vertebrates. PNAS 101
adrenomedullin and calcitonin gene-related peptide receptors. Peptides 28 1638–1643.
1104–1112. Yang B & Ferguson AV 2003 Adrenomedullin influences dissociated rat area
Katafuchi T & Minamino N 2004 Structure and biological properties of three postrema neurons. Regulatory Peptides 112 9–17.
calcitonin receptor-stimulating peptides, novel members of the calcitonin Yang JH, Jia YX, Pan CS, Zhao J, Ouyang M, Yang J, Chang JK, Tang CS &
gene-related peptide family. Peptides 25 2039–2045.
Qi YF 2005 Effects of intermedin 1-53 on cardiac function and
Katafuchi T, Kikumoto K, Hamano K, Kangawa K, Matsuo H & Minamino N
ischemia/reperfusion injury in isolated rat hearts. Biochemical and Biophysical
2003 Calcitonin receptor-stimulating peptide, a new member of the calcitonin
Research Communications 327 713–719.
gene-related peptide family – its isolation from porcine brain, structure, tissue
Watanabe TX, Noda Y, Chino N, Nishiuchi Y, Kimura T, Sakakibara S &
distribution, and biological activity. Journal of Biological Chemistry 278
Imai M 1988 Structure-activity relationships of a-human atrial natriuretic
12046–12054.
peptide. European Journal of Pharmacology 147 49–58.
López J & Martı́nez A 2001 Cell and molecular biology of the multifunctional
peptide, adrenomedullin. International Review of Cytology 221 1–92.
Muff R, Born W, Lutz TA & Fischer JA 2004 Biological importance of the
peptides of the calcitonin family as revealed by disruption and transfer of Received in final form 28 January 2008
corresponding genes. Peptides 25 2027–2038.
Nag K, Kato A, Nakada T, Hoshijima K, Mistry AC, Takei Y & Hirose S 2006
Accepted 19 February 2008
Molecular and functional characterization of adrenomedullin receptors in Made available online as an Accepted Preprint
pufferfish. American Journal of Physiology 290 R467–R478. 19 February 2008
Journal of Endocrinology (2008) 197, 391–400 www.endocrinology-journals.org
Downloaded from Bioscientifica.com at 11/11/2021 01:24:26PM
via free accessYou can also read
