| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Medizinische Klinik und Poliklinik (Kardiologie, Angiologie, Pulmologie) Charité, Campus Mitte, 10098 Berlin, Germany; and
* Immundiagnostik AG, 64625 Bensheim, Germany
1Correspondence: Medizinische Klinik und Poliklinik m. S. Kardiologie, Angiologie, Pulmologie, Charité, Campus Mitte, Schumannstr. 20/21, 10098 Berlin, Germany. E-mail: karl.stangl{at}charite.de
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: RLX • congestive • endothelin 1 • angiotensin II • gene expression
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and
ß—which are connected by two disulfide bridges (1)
. The
two human gene forms coding for H1 and H2 relaxin were identified in
the 1980s (2
, 3)
. In women, the H2 gene is expressed in
the corpus luteum, endometrium, placenta, and breast; H1 mRNA has been
found in the placenta only. In men, H1 and H2 expression has until now
been established exclusively in the prostate gland (1)
. In
human plasma, only H2 RLX has been detected (4)
; the
highest levels are measured in pregnancy and during the second phase of
the menstrual cycle (5)
. Whereas the functional effects of
RLX in the physiology of human reproduction have been known for many
years (6
, 7)
, there is recent evidence from experiments
using exogenous RLX in rats that the peptide may also play a role in
the cardiovascular system owing to its vasodilatory, diuretic, and
central hemodynamic effects (4
, 8)
.
Human congestive heart failure (CHF) is characterized by complex
neurohumoral activation associated with the up-regulation of
vasoconstricting and salt-retaining mediators: catecholamines,
angiotensin II (AT-II), endothelin 1 (ET-1), and vasopressin.
Neurohumoral activation also entails the compensatory rise of
vasodilating and natriuretic mediators: atrial and brain natriuretic
peptides as well as adrenomedullin (9)
. Until now, neither
the possible involvement of RLX in these neuroendocrine alterations nor
its expression in the human cardiovascular system had been
investigated, although the hormone may represent a novel candidate in
view of the above-mentioned cardiovascular actions observed in animal
experiments.
In the present study, we present clinical and experimental data that elucidate the role of RLX as a player in human heart failure.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Group 1: severe CHF
Fourteen CHF patients in functional class NYHA IV were enrolled
(five females, nine males; mean age±SD, 58±7 years, mean
left ventricular ejection fraction [LVEF]±SD, 19±3%).
Seven patients suffered from CHF secondary to ischemic heart disease
(IHD) and seven demonstrated dilated cardiomyopathy (DCM). Inclusion
criteria were:
1) Need for intensive care treatment for acute left heart decompensation (orthopnea, signs of severe pulmonary congestion). Acute ischemia was excluded on the basis of electrocardiography, as well as kinetics of enzymes (troponins, creatin kinase) and myoglobin.
2) PCWP > 25 mmHg, CI < 2.5 l/min/m2.
Group 2: moderate CHF
CHF patients were assigned to the moderate CHF group
(n=13, four females, nine males; age, 54±7 years; LVEF,
27±4%) if they were 1) in NYHA class II and displayed
2) a PCWP < 20 mmHg and a CI > 2.5
l/min/m2. Individuals were recruited among
patients undergoing hemodynamic evaluation before coronary bypass
surgery and/or partial ventriculectomy. Six patients suffered from CHF
secondary to IHD and seven demonstrated DCM.
We treated Group 1 and 2 patients over 12 h with sodium nitroprusside at a dosage that decreased systemic vascular resistance to 800-1000 dyn/s/cm-5 without lowering mean arterial blood pressure below 60 mm Hg. Intravenous (i.v.) furosemide was given according to clinical requirements; i.v. nitrates, catecholamines, and phosphodiesterase inhibitors were not administered. All CHF patients were on stable oral therapy: all on ACE inhibitors and diuretics; 18 on nitrates; 15 on ß-blockers; and 10 on digitalis.
Group 3: controls
Thirteen controls (four females, nine males; age, 55±4 years;
LVEF, 65±5%) were recruited among individuals who were undergoing
cardiac catheterization for suspected coronary artery disease and in
whom no structural cardiovascular disease was detected.
Instrumentation and blood sample collection
In all groups, catheters were positioned in the pulmonary artery
(PA) (Swan-Ganz), the coronary sinus (CS), and the left ventricle (LV)
under fluoroscopic control. Blood sampling and hemodynamic measurements
were carried out at baseline in all groups and at 2, 4, 6, and 12 h after initiation of therapy in Groups 1 and 2. In all groups, an
additional sample of venous blood was drawn 48 h after initiation
of therapy and catheterization. At baseline, arterial blood (A) was
simultaneously drawn from LV and from the radial artery to establish
comparability of these sites. During treatment, arterial blood was
drawn from the radial artery. In addition, blood was taken from PA, CS,
and the antecubital vein (V). In Groups 1 and 2, the left ventricular
catheter and the CS catheter were removed for safety after baseline
measurement and after 4 h, respectively. Arterial and pulmonary
arterial catheters were removed 12–14 h after initiation of therapy.
MAP during treatment was measured invasively in the radial artery.
Heart rate, MAP, and pulmonary arterial pressure (PAP) were
continuously monitored and cardiac output (CO) was determined by
thermodilution.
Exclusion criteria were renal failure with creatinin level over 250 µM, need of i.v. treatment other than SNP and furosemide, severe noncardiovascular systemic diseases, or primary pulmonary hypertension.
This study was approved by the local Ethics Committee and was conducted according to Declaration of Helsinki principles. Before the study, all patients gave their informed, written consent.
Determination of plasma relaxin
RLX was determined by use of a commercial ELISA kit
(Immundiagnostik, Bensheim, Germany). The polyclonal antibody was
raised in rabbits. The kit had a detection limit of 0.40 pg/ml that was
calculated from the mean optical density of the zero standard (measured
in duplicate) plus 2 standard deviations. The intra-assay coefficient
of variation is 9.6% (n=18, at 15 pg/ml) and the interassay
coefficient of variation is 10.2% (n=12, at 15 pg/ml). The
kit is highly selective for human RLX, with cross-reactivity measuring
100% for the H1 form and 100% for the H2 form. Cross-reactivity
against insulin, insulin-like growth factors, LH, FSH, and prolactin is
less than 0.01%.
Determination of ET-1
ET-1 concentrations in plasma and in supernatant of cultured
cells were detected as described elsewhere (10)
.
RNA analysis
We determined expression of prepro-RLX and prohormone convertase
1 (PC-1) mRNA. The latter, also referred to as PC-1/3, has been
reported to process proRLX (11)
. We used the following
tissue samples.
Left ventricles
Control samples were human total left ventricular RNA from
healthy individuals (n=4) (Invitrogen, San Diego, CA) and a
sample from one donor heart that was not transplanted. Failing
myocardium was from patients with DCM (n=5) and IHD
(n=3) who underwent partial ventriculectomy or
transplantation.
Right atria and vessels
Samples from right atrium (n=8), mammary artery
(n=6), and saphenous vein (n=6) were obtained
from patients with normal heart function; vascular specimens
(n=4 each) were also taken from CHF patients. All these
patients underwent bypass surgery. We used samples from dilated right
atria of DCM patients (n=4) who underwent partial
ventriculectomy. In all patients, heart function had been assessed by
preoperative catheterization.
The use of human tissue samples in this study was approved by the local Ethics Committee and was conducted according to Declaration of Helsinki principles. In all cases except for the control left ventricular samples (see above), we had the informed, written consent of the respective patients.
Tissue samples were placed in liquid nitrogen and stored at -70°C
for subsequent RNA isolation. For total RNA isolation according to the
method described by Chomczynski and Sacchi (12)
, tissues
were homogenized, after 20 min incubation at 30°C, in 1.7 ml TRIZOL
Reagent (Life Technologies, Gaithersburg, MD) per 50 mg tissue.
Extraction of RNA was performed by addition of 0.3 ml chloroform per 1
ml TRIZOL, incubation at 30°C for 3 min, and subsequent
centrifugation (12,000 rpm, 30 min, 4°C). RNA in the aqueous phase
was precipitated by addition of 0.8 ml isopropyl alcohol per 1 ml
TRIZOL. After centrifugation (12,000 rpm, 30 min, 4°C), the pellet
obtained was dissolved in diethylpyrocarbonate-treated water.
Total RNA (2 µg) was reverse transcribed using avian myeloblastosis
virus reverse transcriptase and dT15 primers
according to the manufacturer’s instructions (First Strand cDNA
Synthesis Kit, Boehringer Mannheim, Mannheim, Germany). PCR
amplification of single-stranded cDNA was then performed using primer
pairs specific for H1 and H2 prepro-RLX (H1 and H2 5' primer: 5' TCT
GTT TAC TAC TGA ACC AAT TT 3'; H1 3' primer: 5' CTC AAA CAG TGC CAC GTA
GGG TCG 3'; H2 3' primer: 5' ATT AGC CAA TGC ACT GTA GAG TTG 3') (TIB
MOLBIOL), PC-1 (5' primer: 5' GTG AAC TGG AAG CTG ATT TTG CAC G 3'; 3'
primer: 5' CAT AAG AAT TCA TGA CAA AAC AAC C 3'), and for human GAPDH
(5' primer: 5' TGA AGG TCG GAG TCA ACG GAT TTG GT 3'; 3' primer: 5' CAT
GTG GGC CAT GAG GTC CAC CAC 3') (Clontech, Palo Alto, CA). The primer
sequences for H1 and H2 prepro-RLX were selected according to the work
published by Gunnersen and co-workers (13)
. The common 5'
primer is directed against the B chain (bases 35–57 of GenBank
accession number E00220/00219). The specific 3' primers for H1
(accession E00220) and H2 (accession E00219) bind to the A chain (bases
484–507).
Southern blot hybridization was conducted for quantitation of the
amplified sequences. PCR products were separated on 2% agarose gels,
blotted onto nylon membranes (Hybond N, Amersham, Arlington Heights,
IL), hybridized using radioactively labeled oligos specific for H1 or
H2, and directed against the C peptide (bases 249–273 of GenBank
accessions E00220 and E00219) (H1: 5' AAT GAA TTC CAA CAT GAT AAT TAT A
3'; H2: 5' AAC AAA TTC TGA CAT CAT ATT TAT G 3', sequences according to
ref 13
) (Easy Hyb, Boehringer Mannheim). Finally,
autoradiography was performed and autoradiographs were quantified by
use of the ImageMaster 1D Prime software (Pharmacia Biotech, Peapack,
NJ). All data were normalized to GAPDH mRNA expression.
Cloning and sequencing of the different RLX mRNA forms
We used the Eukaryotic TA Cloning Kit (pCR 3.1 vector,
Invitrogen) and the Standard Cycle Sequencing Kit (Applied Biosystems,
Foster City, CA) according to the manufacturer’s instructions.
Immunostaining and immunofluorescence
Sections (5 µm thick) were cut from paraffin-embedded left
ventricular tissues and placed on gelatin-coated glass slides. As
primary antibody, we used the polyclonal antibody described for the RLX
ELISA. Tissues were incubated with the primary antibody (dilution
1:300) for 24 h (4°C); the secondary goat anti-rabbit antibody
and the tertiary reagent were left on the tissues for 1 h at room
temperature. We used 3,3'-diaminobenzidine tetrachloride as chromogen.
In addition, we performed hematoxylin and eosin counterstaining to
assign the RLX immunoreactivity to certain cell types.
We used freeze sections (5 µm) to detect RLX immunoreactivity in left ventricles, right atria, and vessels. The secondary goat anti-rabbit antibody was labeled with the fluorochrome cyanin-3.
Method specificity was tested by substitution of the primary antibody with serum from a nonimmunized rabbit (dilution 1:300).
Western blot analysis
Tissue samples (50 mg) were lysed at 90°C over 30 min in 100
µl diethylpyrocarbonate-treated water plus 100 µl SDS buffer (125
mM Tris-Cl, 20% glycerin, 6% SDS, 10% ß-mercaptoethanol, 0.02%
bromphenol blue). The homogenates were centrifuged at 10.000 rpm over
15 min. Ten microliters of the supernatants were separated on a 17.5%
SDS-PAGE gel. We then blotted separated proteins to polyvinylidene
difluoride membranes (Millipore, Bedford, MA) (300 mA, 1 h). These
membranes were blocked over 2 h at room temperature (5% dried
milk, 0.05% gelatin, 1% BSA, 0.02% Tween 20, 1x TBS [50 mM Tris,
150 mM NaCl]). Membranes were subsequently incubated overnight at
4°C with the primary antibody (dilution 1:200 for RLX and 1:500 for
-actin). As primary antibodies, we used the polyclonal RLX-specific
antibody described for the RLX ELISA and a monoclonal mouse anti-actin
antibody. After washing twice for 5 min and twice for 15 min with 1x
TBS plus 0.02% Tween 20, membranes were incubated over 2 h at
room temperature with the secondary antibody (peroxidase-conjugated
mouse anti-rabbit or goat anti-mouse IgG). Finally, the immunoreactive
bands were visualized with the BM Chemoluminescence system (Boehringer
Mannheim).
The exogenous H1 RLX used for the positive controls was generously provided by Dr. J. Wade from the Howard Florey Institute (Melbourne, Australia); H2 RLX was a gift from Dr. W. Voelter (University of Tuebingen).
Experiments in isolated rat hearts
Isolated Langendorff hearts of male Wistar rats (180–250 g body
weight) were perfused over 2 h in constant pressure mode with
modified Krebs-Henseleit buffer (37°C; pH 7.35–7.45; composition in
mM: NaCl 116, KCl 4.0, MgSO4 1.2,
KH2PO4 1.2,
NaHCO3 25, CaCl2 2.5,
glucose 10, HEPES 6; equilibration with 95% O2
and 5% CO2). A left ventricular enddiastolic
pressure (LVEDP) of either 5 mm Hg (controls) or 25 mm Hg
(n=3 for each group) was adjusted by means of a fluid-filled
latex balloon. In another subset of these experiments (n=3
for each group), we adjusted the mean right or left atrial pressure by
means of a balloon to either normal (5 mm Hg) or elevated values (25 mm
Hg). Subsequently, the respective tissues—free left ventricular or
right/left atrial wall—were rapidly frozen in liquid nitrogen and
stored at -70°C for determination of RLX mRNA. We performed total
RNA isolation and reverse transcription as described above. PCR was
performed using primers for rat prepro-RLX. The 5' primer was directed
against bases 200–222 of the sequence published in ref 14
(GenBank accession number V01264) (5' GGA AGA CTG GCT TTG AGC CAG GA
3'), and the 3' primer binds to bases 608–630 (5' CCG GGT TGC GCA CCA
TTA GCT CC 3') (TIB MOLBIOL).
Cell culture
Bovine pulmonary artery endothelial cells (CPAE, No. CCL-209,
American Type Culture Collection, Rockville, MD), passage 6, were grown
in MEM (Life Technologies) supplemented with 1.5 g/l sodium
bicarbonate, 0.11 g/l sodium pyruvate, 100 U/ml penicillin, 100 µg/ml
streptomycin, and 5% FCS in a humidified 5% CO2
atmosphere. Cells were grown to subconfluence for investigation of the
AT-II-induced ET-1 secretion, but were used at complete confluence for
the flow chamber experiments. The exogenous H2 RLX used in these
experiments was a gift from Dr. W. Voelter (University of Tuebingen).
Flow chamber
The rectangular laminar flow chamber (6.0 cm length, 4.5 cm
width) was perfused with a circulating volume of 300 ml cell culture
medium gassed with 95% air and 5% carbon dioxide over 16 h.
Shear stress 
(in dyn/cm2) was adjusted
according to the following formulas for a Newtonian fluid
(15)
:
 |
 |
, viscosity in dyn s/cm2; Q, laminar
flow in cm3/s; B, chamber width in cm; h, chamber
height in cm; 
p, pressure gradient across the chamber in
dyn/cm2; L, chamber length in cm). The
flow-induced pressure gradient across the chamber was measured as the
difference between inlet and outlet pressure and kept constant for
varying values of by adopting the following procedure: to achieve a
change from 1 to 2,
we adjusted the chamber height h by
2/1 and the flow Q by
(2/1)3,
which yields a constant p and a linear rise of with h.
Data analysis
Data are presented as mean ± SE unless
otherwise indicated. An error probability of P < 0.05
was regarded as significant.
Values for unpaired data (hemodynamic parameters, cell culture
experiments) were compared using the Kruskal-Wallis ANOVA on ranks or
the Mann-Whitney rank sum test. Differences between groups over time
(hemodynamics and peptide levels) were analyzed with a nonparametric
ANOVA for repeated measures (16)
. In each case, a
multiple-comparison procedure with Bonferroni-Holm adjustment of
P was carried out after global testing (17)
.
Regression analyses were performed using commercial software for linear and nonlinear regression (SigmaStat, Jandel Scientific, Corte Madera, CA).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Depending on the sites of measurement (left ventricle, pulmonary
artery, coronary sinus, or antecubital vein), plasma levels of RLX in
moderate CHF (Fig. 1
A) were four- to sixfold higher than those determined in the
control patients. Patients with severe CHF were characterized by
drastically elevated pulmonary pressures and pulmonary vascular
resistance, as well as by markedly lower cardiac index than were
patients with moderate CHF (Table 1)
. These patients with severe CHF
displayed RLX levels 2.2- to 2.6-fold those of patients with moderate
CHF and 12- to 16-fold those detected in controls. To identify
hemodynamic determinants of RLX levels, we performed regression
analyses demonstrating closest correlations of RLX with left
ventricular filling pressure (r=0.69, P<0.001)
and cardiac index (r=-0.62, P<0.001).
|
|
Eleven of the 14 patients with severe CHF (i.e., 79%) showed higher
RLX plasma levels in coronary sinus than in left ventricle (Fig. 1B
). This finding indicates net coronary release of RLX in
the majority of patients suffering from severe CHF.
At baseline, simultaneous determination of plasma RLX from the LV and the radial artery proved the reliability of concentrations detected in the radial artery for monitoring the subsequent treatment throughout 12 h [severe CHF: LV, 18.9±3.4; radial artery; 18.5±3.9 pg/ml (NS); moderate CHF: LV, 7.9±1.2; radial artery, 7.6±1.3 pg/ml (NS)].
We subsequently treated both CHF patient groups for 12 h with the
vasodilator sodium nitroprusside. In patients with severe CHF, the
treatment induced significant changes in mean arterial pressure (after
2 h: -23±2% of baseline values), mean pulmonary arterial
pressure (-39±3%), pulmonary capillary wedge pressure (-46±2%),
and systemic (-65±5%) and pulmonary (-52±3%) vascular resistance.
Results for the cardiac index showed a significant increase (+70±3%).
Maximum hemodynamic effects had already been achieved within the first
2 h of treatment. This acute vasodilator therapy evoked a delayed
decrease in plasma RLX that was detectable 48 h after initiation
of therapy (Fig. 1C
). In patients with moderate CHF, only
the decreases in mean arterial pressure (2 h: -30±1%) and systemic
vascular resistance (3 h: -53±3%) were significant. In these
patients vs. patients with severe CHF, plasma concentrations of RLX did
not change over 48 h.
Increased H1 and H2 gene expression in failing atrial and
ventricular tissues
Because RLX gene expression has not until now been described for
human cardiovascular tissues, we sought to identify candidate
cardiovascular sources of RLX. By RT-PCR, we were able to detect both
H1 and H2 mRNA in left ventricles, right atria, mammary arteries, and
saphenous veins from patients demonstrating normal heart function
(Fig. 2
). These findings suggest constitutive H1 and H2 gene expression in
these tissues. Particularly in the vessels, we found a 101 bp larger
specific PCR product, which represents the splicing form of H1 and H2
mRNA reported by Gunnersen and co-workers for human placenta and
prostate gland (13)
. Sequencing of the different PCR
products (H1, H2, and the splicing form) verified their specificity.
|
We next examined whether the occurrence of CHF was accompanied by
elevated RLX gene expression in myocardial and vascular tissues
compared with the control tissues mentioned above (Fig. 2)
. In left
ventricular specimens of failing hearts, we determined pronounced
up-regulation of GAPDH-normalized H1 mRNA (failing hearts, 14.4±2.4
vs. nonfailing hearts, 1.0±0.5 arbitrary units; P<0.05)
and a moderate increase in H2 mRNA (2.6±0.4 vs. 1.0±0.2 units,
P<0.05). H1 expression was increased moderately in right
atria from failing hearts compared with specimens from nonfailing
hearts (2.1±0.4 vs. 1.0±0.2 units, P<0.05). Results for
right atrial H2 mRNA also showed a more pronounced elevation in failing
hearts (3.8±1.0 vs. 1.0±0.1 units, P<0.05). In contrast,
expression of RLX mRNA species apparently did not differ in mammary
arteries and saphenous veins from patients with normal heart function
and from patients with CHF.
Detection of proRLX protein and gene expression of the processing
enzyme
Western blot analysis in right atrial and left ventricular
specimens obtained from nonfailing and failing hearts (Fig. 3
) revealed a band of the size expected for prorelaxin (
18 kDa)
(11)
in all samples. We further proved that exogenous
mature H1 and H2 was well recognized in our setting, but tissue levels
were obviously below the sensitivity of the method applied. The tissue
concentration of prorelaxin (normalized to -actin) was approximately
doubled in failing right atria (2.3±0.3 vs. 1.0±0.3 arbitrary units),
but unchanged in failing left ventricles when compared with control
tissues (0.9±0.1 vs. 1.0±0.1 units).
|
Although expression of PC-1 has been observed before only in human
neuroendocrine tissues (brain, pituitary and adrenal glands, pancreas)
(11
, 18)
, we clearly detected PC-1 mRNA in nonfailing and
failing atrial and ventricular tissues as well as in vessels (Fig. 3
,
data only shown for myocardium). Moreover, we were able to show that
mRNA expression of the RLX-processing enzyme is regulated in CHF: the
transcript levels are uniformly decreased in failing right atria when
compared with nonfailing tissue. In failing left ventricular
myocardium, however, we observed a broad range of PC1 mRNA expression;
in nonfailing left ventricles, PC1 mRNA was expressed uniformly.
Immunolocalization of RLX in myocardial tissue
Immunostaining combined with hematoxylin and eosin counterstaining
of failing left ventricular tissue disclosed RLX immunoreactivity in
myocytes and interstitial cells (Fig. 4
). Myocytes disclosed a discrete granular pattern of staining that
corresponds well with the results reported for the corpus luteum and
endometrium (19)
. Interstitial cells appeared more
homogeneously stained.
|
The occurrence of RLX-like immunoreactivity was also confirmed by immunofluorescence in freeze sections. A granular fluorescence pattern was observed not only in the myocardium, but also in arterial and venous sections obtained from patients with CHF secondary to ischemic heart disease. In the vessels, however, the fluorescence usually appeared less evenly distributed.
Up-regulation of RLX expression by elevated filling pressure
We tested the hypothesis that myocardial dilatation corresponding
to elevated filling pressures may constitute a key event in the
regulation of RLX (Fig. 5
). Baseline values and time course of heart rate and coronary flow did
not differ between these groups (data not shown). We established that
an elevated LVEDP caused marked up-regulation of RLX gene expression
compared with normal values. In contrast, elevation of both right and
left atrial pressures had no effect on the expression of RLX mRNA (data
not shown for left atria).
|
Sequencing of the PCR product verified that rat prepro-RLX cDNA had been amplified.
Relaxin inhibits stimulated ET-1 secretion
We recently demonstrated that pulmonary circulation represents a
major site of ET-1 net release in patients with severe CHF
(10)
. ET-1, in turn, represents one of the most powerful
mediators of CHF progression (20)
. We therefore
investigated whether RLX may affect pulmonary endothelial
mechano-transduction leading to enhanced ET-1 expression, since such an
effect would suggest a relevant neurohumoral role for RLX in the course
of CHF. In a flow chamber model, we subjected pulmonary artery
endothelial cells to hemodynamic conditions that mimic the in vivo
situation of left ventricular failure with consequent pulmonary
congestion [low arterial shear stress, =18
dyn/cm2, and high downstream pressure at chamber
outlet (‘pulmonary wedge pressure’), P=30 mm Hg]. Here we
found a doubled ET-1 secretion over 16 h compared with normal
hemodynamic constellations (=30 or 50 dyn/cm2,
P=10 mm Hg) (Fig. 6
A). This increase in peptide secretion corresponded to
elevated preproET-1, endothelin-converting enzyme-1 (ECE-1), and ECE-2
mRNA levels (data not shown). At 5 nM, exogenous H2 RLX completely
suppressed this hemodynamically induced increase in ET-1 secretion
(Fig. 6A
), which can be ascribed to suppression of ET-1 gene
expression (data not shown). The activity of human RLX in bovine
endothelial cells also suggests that these cells express a functional
RLX receptor. This RLX–receptor interaction apparently is not
species-restricted, which has also been reported by others
(1)
.
|
In additional experiments, we tested the effects of exogenous H2 RLX on
the AT-II-stimulated ET-1 secretion in pulmonary artery endothelial
cells. This AT-II-induced ET-1 secretion entails a well-established
neurohumoral mechanism contributing to the progression of CHF
(21)
. Whereas ET-1 release is approximately doubled over
4 h and 8 h in the presence of 100 nM AT-II (which
corresponds to elevated levels of mRNA encoding preproET-1 and
ET-converting enzymes 1 and 2), 10 nM of RLX profoundly inhibits this
increase in ET-1 secretion (Fig. 6B
). At the functional
level, selective blockade of ETB receptors
(representing the clearance site for ET-1; ref 22
) by
A-192621 completely prevents the inhibitory action of RLX-2 vis-a-vis
ET-1. This phenomenon suggests an ETB
receptor-mediated mechanism of the RLX effect, which was confirmed by
detection of a massive increase in ETB receptor
gene expression in the presence of AT-II plus RLX (Fig. 6C
).
Altogether, RLX exhibits a strong potential for in vitro suppression of stimulation in the ET-1 system. This finding led us to examine the in vivo correlation between plasma ET-1 and RLX in our patients.
Inverse correlation between plasma ET-1 and RLX in severe heart
failure
In Fig. 6D, we
depict the significant inverse
correlation between circulating ET-1 and RLX in tested patients
suffering from severe CHF (quadratic regression, r=0.81,
P=0.008; linear regression, r=0.71,
P=0.013). Thus, individuals demonstrating the highest RLX
plasma levels manifest the lowest circulating ET-1 concentrations in
this group of patients. In patients with moderate CHF and in controls,
we were not able to establish such a correlation (data not shown).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 7)
, the polypeptide RLX has attracted more general attention
since the early 1990s, when new studies revealed the remarkable
pleiotropy of this mediator in rodents (4
, 8)
. In the
present study, we offer the first evidence that 1) RLX is
constitutively expressed in human cardiovascular tissues and
2) may play a relevant functional role in human heart
failure. This evidence is based on the following. Plasma levels of RLX
in patients increase profoundly with the severity of CHF, and they
respond to acute hemodynamic improvement in the course of intensive
care therapy. In severe CHF, we were able to detect net coronary
release of RLX in more than 75% of the patients studied, which
suggests the heart is a relevant source of elevated circulating RLX in
human CHF. Moreover, we show for the first time that H1 and H2 gene
expression can be detected in the heart and vessels among patients with
preserved heart function as well as in patients suffering from CHF. The
occurrence of CHF, however, is accompanied by significant elevation in
myocardial H1 and H2 mRNA levels. Western blot analysis and
immunostaining revealed translation of RLX mRNA into the RLX
polypeptide in the heart and vessels.
We present novel findings in this study indicating expression of the
proRLX processing enzyme (PC-1) in the heart and vessels. The
differential PC-1 mRNA expression observed in failing atria and
ventricles may explain the distinct pattern of proRLX protein levels:
with H1 and H2 transcripts being elevated at both sites, decreased
atrial expression of PC-1 may facilitate proRLX increase, whereas the
nonuniform ventricular regulation of PC-1 may result in varying levels
of proRLX. From immunostaining, myocytes and interstitial cells appear
as the myocardial sources of RLX. At the functional level, myocardial
dilation due to increased left ventricular filling pressure constitutes
a regulatory event for RLX gene expression. Moreover, exogenous RLX was
shown to effectively suppress the pulmonary endothelial ET-1 response
to hemodynamic and humoral (AT-II) stimuli. RLX modulates AT-II
effects, indicated by the increase of endothelial
ETB receptor expression in the presence of AT-II
plus RLX. The potential impact of these in vitro findings was
accentuated by the data documented in our patients with severe CHF:
circulating RLX shows a close inverse correlation with plasma ET-1, the
most potent vasoconstrictor in heart failure and a powerful mediator of
salt-retention and myocardial remodeling (20)
.
From our findings and from the body of data obtained by
investigating the effects of exogenous RLX in rodents, we may predict
that the peptide chiefly exerts compensatory effects in the course of
CHF: i.e., vasodilation (8)
, diuresis caused by
attenuation of the renal vascular response to AT-II (23)
,
stimulation of atrial natriuretic peptide (24)
, collagen
matrix degradation (25)
, prevention of coronary thrombotic
events by up-regulating tissue plasminogen activator (26)
modulation of AT II effects, and suppression of the ET-1 system.
In conclusion, we have identified a potential new player in human heart failure. Relaxin may prove valuable in assessing the prognosis of heart failure and represents a potential target for future therapeutic strategies.
|
ACKNOWLEDGMENTS |
|---|
Received for publication February 22, 2001.
Revision received June 4, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  Q. Xu, E. D. Lekgabe, X.-M. Gao, Z. Ming, G. W. Tregear, A. M. Dart, R. A. D. Bathgate, C. S. Samuel, and X.-J. Du Endogenous Relaxin Does Not Affect Chronic Pressure Overload-Induced Cardiac Hypertrophy and Fibrosis Endocrinology, February 1, 2008; 149(2): 476 - 482. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Dschietzig, C. Richter, L. Asswad, G. Baumann, and K. Stangl Hypoxic Induction of Receptor Activity-Modifying Protein 2 Alters Regulation of Pulmonary Endothelin-1 by Adrenomedullin: Induction under Normoxia Versus Inhibition under Hypoxia J. Pharmacol. Exp. Ther., April 1, 2007; 321(1): 409 - 419. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X.-l. Moore, S.-l. Tan, C.-y. Lo, L. Fang, Y.-D. Su, X.-M. Gao, E. A. Woodcock, R. J. Summers, G. W. Tregear, R. A. D. Bathgate, et al. Relaxin Antagonizes Hypertrophy and Apoptosis in Neonatal Rat Cardiomyocytes Endocrinology, April 1, 2007; 148(4): 1582 - 1589. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Novak, L. J. Parry, J. E. Matthews, L. J. Kerchner, K. Indovina, K. Hanley-Yanez, K. D. Doty, D. O. Debrah, S. G. Shroff, and K. P. Conrad Evidence for local relaxin ligand-receptor expression and function in arteries FASEB J, November 1, 2006; 20(13): 2352 - 2362. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Hombach-Klonisch, J. Bialek, B. Trojanowicz, E. Weber, H.-J. Holzhausen, J. D. Silvertown, A. J. Summerlee, H. Dralle, C. Hoang-Vu, and T. Klonisch Relaxin Enhances the Oncogenic Potential of Human Thyroid Carcinoma Cells Am. J. Pathol., August 1, 2006; 169(2): 617 - 632. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. S. Samuel Relaxin: Antifibrotic Properties and Effects in Models of Disease Clin. Med. Res., November 1, 2005; 3(4): 241 - 249. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. B. Yap, D. Mukerjee, P. M. Timms, H. Ashrafian, and J. G. Coghlan Natriuretic Peptides, Respiratory Disease, and the Right Heart Chest, October 1, 2004; 126(4): 1330 - 1336. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. S. Samuel, E. N. Unemori, I. Mookerjee, R. A. D. Bathgate, S. L. Layfield, J. Mak, G. W. Tregear, and X.-J. Du Relaxin Modulates Cardiac Fibroblast Proliferation, Differentiation, and Collagen Production and Reverses Cardiac Fibrosis in Vivo Endocrinology, September 1, 2004; 145(9): 4125 - 4133. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. P. Goetze Biochemistry of Pro-B-Type Natriuretic Peptide-Derived Peptides: The Endocrine Heart Revisited Clin. Chem., September 1, 2004; 50(9): 1503 - 1510. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. P. Conrad and J. Novak Emerging role of relaxin in renal and cardiovascular function Am J Physiol Regulatory Integrative Comp Physiol, August 1, 2004; 287(2): R250 - R261. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Heringlake, C. Heide, L. Bahlmann, W. Eichler, H. Pagel, P. Schmucker, R. Wergeland, F. P. Armbruster, and S. Klaus Effects of tilting and volume loading on plasma levels and urinary excretion of relaxin, NT-pro-ANP, and NT-pro-BNP in male volunteers J Appl Physiol, July 1, 2004; 97(1): 173 - 179. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Hocher, R. Ziebig, R. Krause, G. Asmus, H.-H Neumayer, L. Liefeldt, and J.-P. Stasch Relaxin Is an Independent Risk Factor Predicting Death in Male Patients With End-Stage Kidney Disease Circulation, May 18, 2004; 109(19): 2266 - 2268. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. D. Sherwood Relaxin's Physiological Roles and Other Diverse Actions Endocr. Rev., April 1, 2004; 25(2): 205 - 234. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. P. Goetze, J. Kastrup, and J. F Rehfeld The paradox of increased natriuretic hormones in congestive heart failure patients: Does the endocrine heart also fail in heart failure? Eur. Heart J., August 2, 2003; 24(16): 1471 - 1472. [Full Text] [PDF]
|
|
|
|
 |
|
 C Fisher, C Berry, L Blue, J J Morton, and J McMurray N-terminal pro B type natriuretic peptide, but not the new putative cardiac hormone relaxin, predicts prognosis in patients with chronic heart failure Heart, August 1, 2003; 89(8): 879 - 881. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C Fisher, S Al-Benna, A Kirk, J J Morton, and J J V McMurray Transcardiac and transpulmonary gradients in the putative new cardiovascular hormone relaxin Heart, July 1, 2003; 89(7): 789 - 790. [Full Text] [PDF]
|
|
|
|
 |
|
 B. T. Nguyen, L. Yang, B. M. Sanborn, and C. W. Dessauer Phosphoinositide 3-Kinase Activity Is Required for Biphasic Stimulation of Cyclic Adenosine 3',5'-Monophosphate by Relaxin Mol. Endocrinol., June 1, 2003; 17(6): 1075 - 1084. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X.-J. Du, C. S Samuel, X.-M. Gao, L. Zhao, L. J Parry, and G. W Tregear Increased myocardial collagen and ventricular diastolic dysfunction in relaxin deficient mice: a gender-specific phenotype Cardiovasc Res, February 1, 2003; 57(2): 395 - 404. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Dschietzig, C. Bartsch, C. Richter, M. Laule, G. Baumann, and K. Stangl Relaxin, a Pregnancy Hormone, Is a Functional Endothelin-1 Antagonist: Attenuation of Endothelin-1-Mediated Vasoconstriction by Stimulation of Endothelin Type-B Receptor Expression via ERK-1/2 and Nuclear Factor-{kappa}B Circ. Res., January 10, 2003; 92(1): 32 - 40. [Abstract] [Full Text] [PDF]
|
|
|
|
moderate CHF vs. controls.
B) Eleven of 14 patients with severe CHF (i.e., 79%)
revealed coronary net release of RLX, i.e., plasma levels are higher in
CS than in LV. C) Course of circulating venous RLX during
and after 12 h vasodilator therapy with sodium nitroprusside.
P < 0.05; #vs. baseline.
) and in
interstitial cells (*) in left ventricular tissue from a patient
demonstrating dilated cardiomyopathy. Magnification: x80
(A), x66
(B), and x40 (C).
D, E) Specificity was verified by substitution of the primary
antibody with serum from a nonimmunized rabbit. Magnification: x66
(D) and x40 (C). F)
Freeze sections, RLX immunofluorescence in myocardial and vascular
specimens from patients demonstrating CHF secondary to ischemic heart
disease. Magnification: x80 (left ventricle and right atrium) and x40
(mammary artery and saphenous vein).
