Full text of DOI:10.1097/00003246-200105000-00031

Pulmonary adrenomedullin counteracts deterioration of coronary
flow and myocardial performance evoked by pulmonary
endothelins in experimental acute respiratory distress syndrome
Karl Stangl, MD; Michael Laule, MD; Christoph Richter, MD; Verena Stangl, MD;
Jonas Koch, MD; Öndar Göktas, MD; Gert Baumann, MD; Thomas Dschietzig, MD
Objectives: We recently showed that pulmonary endothelins and phosphoramidon prevented this effect, whereas blockade of
may affect coronary circulation under various experimental and endothelin type B and adrenomedullin-binding receptors had no
clinical conditions. Here, we investigated the effect of pulmonary effect. After endotoxin challenge, coronary flow significantly in-
mediators on coronary tone in experimental acute respiratory creased to 110 ؎ 2%. This response was augmented by BQ-123
distress syndrome. We focused particularly on pulmonary endo- (124 ؎ 2%) and phosphoramidon (123 ؎ 3%); A-192621 had no
thelin-1, a major vasoconstrictor in acute respiratory distress effect. Application of hCGRP(8-37) and hAM(22-52) significantly
syndrome, and on adrenomedullin, a potent vasodilator that is decreased coronary flow to 81 ؎ 3% and 88 ؎ 2%, respectively.
up-regulated by inflammatory stimuli.
Design: Controlled experiment that used isolated organs.
Setting: Experimental laboratory.
Flow decrease after blockade of both adrenomedullin-binding
receptors (73 ؎ 2%) significantly deteriorated peak left ventric-
ular pressure, to 82 ؎ 6% of baseline; rate of pressure increase,
to 81 ؎ 5%; and rate of pressure decline, to 77 ؎ 6%. Endotoxin
Subjects: Wistar rats.
Interventions: The saline effluent from an isolated lung was pretreatment elevated pulmonary venous big endothelin-1 (three-
used to serially perfuse the coronary vessels of an isolated heart. fold), endothelin-1 (two-fold), and adrenomedullin (five-fold).
We compared serial perfusion after 2-hr pretreatment of lungs
Conclusion: In experimental acute respiratory distress syn-
with vehicle or endotoxin (50 ␮g/mL), and we used the following drome, pulmonary adrenomedullin—via calcitonin gene-related
drugs to elucidate the coronary response observed: the endothelin peptide type-1 receptor and adrenomedullin receptor—outweighs
type A receptor antagonist BQ-123 (2 ␮M), the endothelin type B the coronary vasoconstrictor impact of pulmonary big endothe-
antagonist A-192621 (500 nM), the endothelin-converting enzyme lin-1 exerted via endothelin type A receptors after conversion to
inhibitor phosphoramidon (50 ␮M), the calcitonin gene-related mature endothelin-1. The consequence is prevention of flow-
peptide type-1 receptor antagonist hCGRP(8-37) (2 ␮M), and the related deterioration of myocardial performance. (Crit Care Med
adrenomedullin receptor antagonist hAM(22-52) (200 nM) (n ؍ 6 2001; 29:1027–1032)
each).
KEY WORDS: coronary circulation; endothelins; adrenomedullin;
Measurements and Main Results: In controls, serial perfusion pulmonary circulation; inflammation; contractile function
decreased coronary flow to 87 ؎ 3% of baseline (p < .05). BQ-123
ung tissue in general and pul- nary ETs—big ET-1 and ET-1—exert hand, the previously mentioned up-
monary vasculature in particu- such an influence: namely, elevation of regulation of pulmonary AM system by
lar produce a variety of vasoac- coronary tone under basal conditions inflammatory stimuli might counteract
L
tive mediators, including (8), in experimental pulmonary embo- ET-1 effects. AM—a 52 amino acid-
arachidonic acid metabolites (1); endo- lism (9), and in severe human heart peptide originally isolated from human
thelium-derived relaxing factor (nitric failure (10).
pheochromocytoma (14)—is produced
oxide) (2); atrial (3), brain (4), and C-type
The manner in which this remote pul- by endothelial cells, vascular smooth
natriuretic peptides (5); endothelin monary effect on the coronary vascular muscle cells, myocytes, and various
ET)-1 (6); and adrenomedullin (AM) (7). bed is altered under conditions of acute other cell types. Apart from its vaso-
(
Through release of these mediators
respiratory distress syndrome (ARDS) is a dilatory action, AM shows natriuretic
matter of pathophysiological interest, be- as well as antiproliferative effects,
cause left ventricular failure is a common and it is currently regarded as one of
complication of ARDS. Different inflam- the most important counterregulatory
matory stimuli have been reported to up- mediators in human heart failure
regulate vasoactive mediators in the (15).
into pulmonary circulation, the lungs
may affect coronary vascular tone. We
have recently demonstrated in experi-
mental and clinical studies that pulmo-
lungs, including the vasoconstrictor ET-1
In the present experimental study we
(
11) and the vasodilatory peptide AM (12). addressed how endotoxin-treated lungs
From Medizinische Klinik und Poliklinik I (Kardiolo-
gie), Charité (Campus Mitte), Humboldt-Universität zu
Berlin, Germany.
Besides thromboxane, ET-1 constitutes a alter coronary tone, with particular at-
major mediator of pulmonary hyperten- tention to the contrary actions of pulmo-
Copyright © 2001 by Lippincott Williams & Wilkins sion during ARDS (13). On the other nary ETs and pulmonary AM.
Crit Care Med 2001 Vol. 29, No. 5
1027

MATERIALS AND METHODS
mg/hr), and no signs of hemorrhage, edema, for ET
or atelectasis. selective antagonist A-192621 (500 nM), char-
The apparatus used in our investigation acterized by potency of agonist values of 8.4
B
receptors (20); the ET receptor-
B
The study conforms to guidelines pub-
lished by the U.S. National Institutes of Health was manufactured by Hugo Sachs Electronics for ETB
receptors and 5.2 for ETA receptors
(March-Hugstetten, Germany). Modification (21); phosphoramidon (50 ␮M), an inhibitor of
Model of Serial Lung-Heart Perfusion. Ow- of the device allows rapid switching from sep- ET-converting enzymes (ECE) with an IC50
(
16).
ϭ
ing to difficulties involved in selectively focus- arate to serial perfusion of heart and lungs. 1 ␮M for ECE-1 and an IC50 ϭ 4 nM for ECE-2
ing on pulmocoronary interactions in a com- With a switch, the amount of pulmonary ve- (22); hCGRP(8-37) (2 ␮M), an antagonist of
plex in vivo model, we have developed a model nous effluent that equals coronary flow is rap- the calcitonin gene-related peptide (CGRP)
of serial lung-heart perfusion. This model has idly conveyed to the heart apparatus, where it type-1 receptor (CGRP-R1) demonstrating a
been described in detail for rabbits (8), and in is oxygenated before becoming perfused potency of agonist value of 7.0 in rat coronary
the present study we used it with the same through the coronary vessels.
arteries (23); and hAM(22-52) (200 nM), an
techniques for Wistar rats (200–300 g body
Experimental Protocol—Serial Perfusion antagonist of the specific AM receptor, the
weight) with the exception that it was neces- Experiments. During the first phase of these IC50 of which is 70 nM in rat mesangial cells
sary for us to use hearts and lungs from dis- experiments, both organs were perfused sepa- (24). The latter two antagonists were chosen
tinct rats because of the 2-hr pretreatment of rately. Initially, the lung was perfused in non- for two reasons: first, because AM elicits its
the lungs (see subsequent description of pro- recirculatory mode and controlled over 30 effects via the CGRP-R1 or the AM receptor,
tocol). We validated in numerous control ex- mins for constant mean pulmonary arterial depending on the tissues being studied, and
periments in rats (n ϭ 20) that no differences pressure and weight. Lung perfusion subse- second, because several reports have indicated
arise on comparison of serial saline lung-heart quently was conducted in recirculatory mode that AM action in rats is not in all cases sen-
perfusions prepared from the same animal over 2 hrs with circulating volume of 80 mL, sitive to hCGRP(8-37) (25, 26). For these two
with those prepared from two different ani- in the presence of either lipopolysaccharide reasons, we administered hCGRP(8-37) and
mals.
(LPS; 50 ␮g/mL) or vehicle alone (controls) (n hAM(22-52) alone and in combination.
Isolated hearts and isolated lungs were pre- ϭ 6 for both). Finally, perfusion mode for the
Drugs were applied as coronary infusion,
pared after intraperitoneal injection of pento- lung was switched back in both groups to which in all cases commenced 15 mins before
barbital sodium (50 mg/kg) with use of stan- nonrecirculatory perfusion with LPS-free the switch from separate to serial perfusion
dard procedures. We perfused the hearts at buffer. Ten minutes after this switch, isolated and which we maintained throughout the ex-
constant pressure of 60 mm Hg with a modi- hearts had already stabilized over 20 mins, and periment.
fied Krebs-Henseleit solution (in mM: NaCl it was possible to conduct serial perfusion over
16, KCl 4.0, MgSO4 1.2, KH PO 1.2, 30 mins in both groups. At the beginning of To compare the results obtained in serial per-
Control Experiments in Isolated Hearts.
1
2
4
NaHCO 25, CaCl 2.5, glucose 10, HEPES 6; serial perfusion, LPS content in pulmonary fusion experiments with effects of exogenous
3
2
3
7.0°C; pH ϭ 7.35–7.40), equilibrated with venous effluent of both groups was Ͻ1 ng/mL vasoactive mediators, and to verify the effec-
9
5% oxygen and 5% CO . Left ventricular (LAL assay; BioWhittaker, Verviers, Belgium). tiveness of the different antagonists, isolated
2
pressures (LVP) were recorded with a fluid- By using this protocol we avoided any possible hearts received 30-min infusions of 5 pM rat
filled latex balloon inserted through the mitral bias of CF by the direct vasodilatory effects of ET-1 and big ET-1, as well as 100 pM rat AM (n
valve and attached to a pressure transducer. exogenous LPS known to occur in rats at ϭ 6 each). Moreover, ET-1 and big ET-1 were
End-diastolic LVP was maintained between 3 concentrations Ն1 ␮g/mL (17).
administered in the presence of BQ-123,
and 6 mm Hg. Balloon pressures were differ-
Experimental ARDS. The LPS challenge in A-192621, and phosphoramidon, and AM was
entiated electronically to yield rates of pres- rats is a well-characterized ARDS model: First, given in presence of hCGRP(8-37) and
sure increase and decrease (dP/dt) and heart rats—in contrast with other species—do not hAM(22-52) (n ϭ 6 for each drug). All antag-
rate (HR). An ultrasonic flow probe (Trans- display relevant pulmonary vascular response onists were applied 15 mins before infusion of
sonic Systems, Fürstenfeldbruck, Germany) to LPS. This phenomenon has been observed vasoactive mediators began. In additional con-
inserted in the aortic perfusion line measured by several investigators in vivo (11) and in trol experiments (n ϭ 3 for each drug), we
the coronary flow (CF). A pressure transducer vitro (18). But, despite this lack of pulmonary verified that none of the drugs interfered with
attached to the aortic perfusion cannula mea- vascular response, rats challenged with LPS in CF or with cardiac contractility over 45 mins.
sured coronary perfusion pressure (CPP). Reg- vivo develop relevant characteristics of ARDS,
Drugs. Rat AM and hCGRP(8-37) were pur-
istration included HR, LVP, left ventricular that is, pulmonary leukocyte accumulation, chased from Bachem (Heidelberg, Germany).
peak dP/dt (LVdP/dtmax and LVdP/dtmin), CF, followed by lung damage and increased per- Rat ET-1, big ET-1, hAM(22-52), BQ-123,
and CPP. Only those hearts were accepted that meability (19). Because the endotoxin-induced phosphoramidon, and salmonella minnesota
achieved a contractile performance of Ͼ80 pulmonary injury is mediated mainly by acti- LPS were from Sigma Chemical (Munich, Ger-
mm Hg LVPmax and Ͼ1500 mm Hg/sec LVdP/ vated neutrophils (19), we documented no many). A-192621 was a gift from Abbott Labs
dtmax and LVdP/dtmin
.
change in permeability in our model of blood- (Abbott, IL).
Determination of Vasoactive Peptides in
Lungs were ventilated with a small-animal free perfusion, which confirms data reported
respirator (60 strokes/min, tidal volume 8–10 by Uhlig et al (18). Hence, the lung model Pulmonary Venous Effluent. The amount of
mL/kg body weight, 1 mm Hg positive end- used in these experiments allows changes in pulmonary venous effluent not conveyed to
expiratory pressure, gas mixture 21% oxygen, pulmonary mediators to be observed without the heart owing to the difference between cor-
5
% CO , and 74% N ). We performed perfu- interference by mechanical damage or altered onary and pulmonary flow was sampled during
2 2
sion with the Krebs-Henseleit buffer described hemodynamics. On the other hand, functional serial perfusion according to the following
previously. Lungs were suspended in a humid- consequences of these alterations can be as- protocol: sample 1, minutes 0–10; sample 2,
ified chamber from a force transducer that sessed only in intact animals and are beyond minutes 11–20; sample 3, minutes 21–30. Af-
monitored changes in lung weight. Within 20 the scope of this study. ter admixture of 0.2 g% bovine serum albu-
mins, pulmonary flow was increased gradually Pharmacological Interventions During Se- mine, samples were rapidly frozen in liquid
and finally maintained at 14 mL/min. Only rial Perfusion. Each of the following drugs was nitrogen and stored at Ϫ70°C for subsequent
those lungs were selected for the present study given to different subgroups (n ϭ 6 for each determination of big ET-1, ET-1, and AM.
that showed constant mean pulmonary arte- subgroup) of both LPS and control prepara- In additional experiments (n ϭ 4 for both
rial pressure (5–8 mm Hg, zero-referenced at tions: BQ-123 (2 ␮M), an ETA receptor- LPS and controls), we excluded the possibility
hilum), constant peak inflation pressure of selective antagonist demonstrating an IC50
ϭ
that preceding recirculatory perfusion of the
7
–10 mm Hg, practically no weight gain (Ͻ50 7.3 nM for ET receptors and an IC ϭ 18 ␮M lungs alters pulmonary venous concentrations
A 50
1
028
Crit Care Med 2001 Vol. 29, No. 5

of ET and AM during serial perfusion. Indeed, Ϯ 119 vs. 1567 Ϯ 133 mm Hg/sec. Also,
adopting the protocol described under Exper- control and LPS lungs exhibited similar
imental Protocol—Serial Perfusion Experi-
mean pulmonary arterial pressure and
ments—with the exception that lung perfu-
weight gain at the end of 2 hrs of recir-
sion was exclusively conducted in
culatory perfusion (controls, 7.3 Ϯ 0.5
nonrecirculatory mode—changed neither big
mm Hg, ϩ10 Ϯ 12 mg; LPS, 7.5 Ϯ 0.6
ET-1, ET-1, nor AM values in pulmonary ve-
mm Hg, Ϫ1 Ϯ 13 mg) and during non-
recirculatory perfusion immediately be-
fore serial perfusion (controls, 7.5 Ϯ 0.5
nous effluent (data not shown).
Radioimmunoassays for Big ET-1 and
ET-1. The radioimmunoassay for rat big ET-1
has been established by Brunner et al. (27), as mm Hg, ϩ9 Ϯ 11 mg; LPS, 7.4 Ϯ 0.8 mm
described previously. To summarize, a poly- Hg, ϩ2 Ϯ 9 mg).
clonal antiserum against big ET-1 (1-39; rat)
was raised in rabbits. Standard solutions of big
ET-1 (1-39; rat) were used at 2–512 pg/assay;
bound radioactivity was separated by using
Serial Perfusion
polyethylene glycol. The detection limit was
Figure 1 shows the coronary flow re-
sponse in the control and the LPS
groups. In controls, CF reached a new
~2 pg per assay tube (~10% displacement of
Figure 1. Coronary flow during serial perfusion of
isolated hearts with effluent of control lungs (n
ϭ 6) and with effluent of lungs pretreated with
the radioactive tracer), and the IC50 value was
5 pg/tube. The cross-reaction of the antiserum
was 16% with big ET-1 fragment 22-39 (bovine),
4
steady state at 87 Ϯ 3% of the baseline lipopolysaccharide (LPS; 50 ␮g/mL) over 2 hrs (n
and zero with rat ET-1, rat atrial natriuretic value after approximately 10 mins. In ϭ 6). Switch from separate to serial perfusion
peptide (1-28), and rat angiotensin II (~0.1% at contrast, CF in the LPS group increased was performed at time zero (baseline). Data are
the IC50 concentration in each case).
to 110 Ϯ 2% of baseline rate. In both given as percentage of baseline coronary flow. #p
groups, neither HR nor contractile pa-
rameters changed significantly compared
with baseline values (data not shown).
whereas during the other interventions
neither HR nor contractility changed sig-
nificantly (data not shown). In additional
The concentration of rat ET-1 was deter-
mined by radioimmunoassay by using a com-
mercial kit (Peninsula, Belmont, CA) as de-
scribed previously (28). The detection limit
was ~0.15 pg/tube; the cross-reactivity of
other ET isomers and big ET-1 in this assay
Ͻ .05, control vs. LPS.
was Ͻ5% and Ͻ37%, respectively, according Pharmacological Interventions
experiments in isolated hearts, we failed
to find that the alterations in mechanical
to the supplier.
During Serial Perfusion
Radioimmunoassay for Adrenomedullin.
Rat AM was determined by the use of a com-
Figure 2 shows CF during the various performance after combined application
mercial radioimmunoassay kit (Phoenix Phar- pharmacological interventions. None of of the CGRP-R1 and AM receptor antag-
maceuticals, Mountain View, CA). This kit (de-
the drugs or drug combinations influ- onists differed significantly from those
enced CF, HR, or contractile parameters evoked by simple flow reduction to cor-
during the 15-min pretreatment immedi- responding values (n ϭ 4) (data not
tection limit 0.3 pg/tube) demonstrates high
selectivity for rat AM(1-50) (100%). With this
kit, cross-reactivity for rat proadrenomedullin
ately before beginning serial perfusion shown).
N-terminal 20 peptide, human amylin, and rat
ET-1 is very low (Ͻ0.1%).
(
data not shown). In controls, the ETA
In both control and LPS groups, BQ-
antagonist BQ-123 and the ECE inhibitor 123 and phosphoramidon were equipo-
phosphoramidon completely inhibited tent. This result indicates that coronary
Statistical Analysis. Data are presented as
mean Ϯ SEM. Baseline values of isolated organs
were compared by using the Kruskal-Wallis the decrease in CF during serial perfu- constriction was mediated by conversion
analysis of variance on ranks. Differences be- sion, whereas A-192621 (data not shown), of pulmonary big ET-1 rather than by
tween groups over time were analyzed by a hCGRP(8-37), hAM(22-52), and the com- direct action of the mature peptide.
nonparametric analysis of variance for re-
bination of the latter two drugs had no
peated measures (29). After global testing, we
effect. In the LPS group, we observed the
performed a multiple-comparison procedure
following coronary responses: Presence
Control Experiments in Isolated
Hearts
with Bonferroni-Holm adjustment of p (30).
An error probability of p Ͻ .05 was regarded as
significant.
of BQ-123 and phosphoramidon signifi-
cantly augmented flow increase, to 124 Ϯ
2
% and 123 Ϯ 3% of baseline rate, re-
In Figure 3, we depict the effects of
spectively, compared with untreated exogenous vasoactive mediators on CF.
preparations. Administration of A-192621
had no effect (data not shown). Applica-
Coronary infusion of ET-1 at a con-
centration of 5 pM significantly decreased
RESULTS
tion of hCGRP(8-37) and hAM(22-52) sig- CF to 84 Ϯ 3% of baseline value; an
nificantly decreased CF, with values re- identical concentration of big ET-1 de-
corded at 81 Ϯ 3% and 88 Ϯ 2%, creased CF to 86 Ϯ 2%. ETA receptor
Serial Perfusion—Baseline
Immediately before beginning serial respectively. Furthermore, combined ap- blockade by BQ-123 inhibited the effects
perfusion, all relevant variables were plication of the latter two drugs resulted of both ET-1 and big ET-1 completely,
nearly identical in hearts of the control in a significantly steeper decrease in CF whereas ECE inhibition by phosphoram-
and the LPS groups: CF, 7.5 Ϯ 0.5 vs. 7.7 compared with the single administra- idon prevented only the big ET-1-induced
Ϯ 0.6 mL/min; HR, 322 Ϯ 15 vs. 315 Ϯ tions: i.e., to 73 Ϯ 2% of baseline. In this decrement in CF. ETB blockade by
2
0 beats/min; LVPmax, 95 Ϯ 7 vs. 93 Ϯ 8 case, LVPmax (82 Ϯ 6% of baseline), A-192621 had no effect (data not shown).
mm Hg; LVdP/dtmax, 2004 Ϯ 123 vs. 2103 LVdP/dtmax (81 Ϯ 5%), and LVdP/dtmin AM at a concentration of 100 pM in-
Ϯ 143 mm Hg/sec; and LVdP/dtmin, 1634 (77 Ϯ 6%) deteriorated significantly, creased CF to 134 Ϯ 4% of baseline rate.
Crit Care Med 2001 Vol. 29, No. 5
1029

both groups, values differed only slightly
at the various time intervals.
Adrenomedullin. At the various time
intervals during serial perfusion, control
lungs released between 9.6 Ϯ 0.8 and
1
0.2 Ϯ 0.7 pM into pulmonary venous
effluent. After LPS treatment, these val-
ues incremented approximately five-fold
and were measured between 48.5 Ϯ 2.7
and 51.2 Ϯ 3.1 pM.
DISCUSSION
We investigated the effect of pulmo-
nary mediators on coronary vascular tone
under conditions of experimental ARDS.
We have recently shown pulmonary ETs
to act—via ETA receptors—as remote
coronary vasoconstrictors (8, 9, 10). In
the present study, we confirmed these
findings in rats and further established
that the basal pulmonary vasoconstrictor
impact—exerted by big ET-1—trans-
forms into a vasodilator effect after pul-
monary LPS challenge. This vasodilation
is primarily caused by elevated pulmo-
nary AM as we were able to demonstrate
by the fact that combined blockade of the
two AM-binding receptors prevented a de-
crease in coronary tone. Correspond-
ingly, we measured increased AM concen-
Figure 2. Coronary flow during serial perfusion of
isolated hearts with (A) effluent of control lungs
and (B) effluent of lungs pretreated with lipopoly-
saccharide (LPS; 50 ␮g/mL) over 2 hrs. A, curve
Figure 3. Coronary flow in isolated hearts in the
presence of exogenous endothelin (ET)-1, big trations in the pulmonary effluent of
ET-1, and adrenomedullin (AM). Data are given LPS-treated lungs.
as percentage of baseline coronary flow; n ϭ 6 for
each group. A, curve 1 is ET-1 (5 pM); 2, big ET-1
The AM in pulmonary effluent most
likely originates from pulmonary vascu-
lar endothelial and smooth muscle cells.
This conclusion is based on the fact that
these cell types recently have been iden-
tified as the main source of the peptide
(31, 32). In addition, enhanced expres-
sion of AM mRNA as well as elevated
peptide synthesis has been reported in
1
is control lungs; 2, the endothelin (ET)A-
selective antagonist BQ-123 (2 ␮M); 3, the ET-
converting enzyme (ECE) inhibitor phosphoram-
idon (50 ␮M); 4, the calcitonin gene-related
peptide type-1 receptor (CGRP-R1) antagonist
hCGRP(8-37) (2 ␮M); 5, the adrenomedullin
(
␮
5 pM); 3, ET-1 plus ETA antagonist BQ-123 (2
M); 4, ET-1 plus ET-converting enzyme inhibi-
tor phosphoramidon (50 ␮M); 5, big ET-1 plus
BQ-123 (2 ␮M); and 6, big ET-1 plus phosphor-
amidon (50 ␮M). B, Curve 1 is AM (100 pM); 2,
AM plus calcitonin gene-related peptide type-1
receptor antagonist hCGRP(8-37) (2 ␮M); 3, AM
plus AM receptor antagonist hAM(22-52) (200
(
AM) receptor antagonist hAM(22-52) (200 nM);
and 6, hCGRP(8-37) plus hAM(22-52). B, curve 1
is LPS lungs; 2, the ET -selective antagonist BQ-
A
1
23 (2 ␮M); 3, the ECE inhibitor phosphorami-
nM); and 4, AM plus hCGRP(8-37) plus hAM(22- vivo (12), as well as after LPS challenge in
2). #p Ͻ .05, vs. untreated application of sub- cultured endothelial and vascular smooth
don (50 ␮M); 4, the CGRP-R1 antagonist
hCGRP(8-37) (2 ␮M); 5, the AM receptor antag-
onist hAM(22-52) (200 nM); 6, hCGRP(8-37) plus
hAM(22-52); and 7, hCGRP(8-37) plus hAM(22-
5
stances; *p Ͻ .05, single application of hCGRP(8-
7) or hAM(22-52) vs. hCGRP(8-37) plus
muscle cells (31, 32).
3
With regard to the receptors involved
in the coronary action of AM, our find-
ings indicate contribution of both the
CGRP-R1 and the AM receptor. To our
knowledge, hAM(22-52) has not yet been
tested in isolated rat hearts, but there are
hAM(22-52).
5
2) plus BQ-123. Six experiments were per-
formed for each group. Switch from separate to
serial perfusion was performed at time zero
Determination of ET and AM in
Pulmonary Venous Effluent
(
baseline). Data are given as percentage of base-
line coronary flow. #p Ͻ .05, vs. untreated con-
trol and untreated LPS; *p Ͻ .05, single applica-
tion of hCGRP(8-37) or hAM(22-52) vs.
hCGRP(8-37) plus hAM(22-52).
Table 1 displays the concentrations several reports in good concurrence with
detected for the different sampling peri- our findings. For example, Entzeroth et
ods during serial perfusion.
Big ET-1 and ET-1. In controls, pul- but not complete blockade, by hCGRP(8-
Both the CGRP-R1 antagonist hCGRP(8- monary venous concentrations of big 37) of the AM-evoked increase in CF in
7) and the AM receptor antagonist ET-1 and ET-1 during serial perfusion isolated rat hearts. Sheykhzade et al. (23)
al. (33) reported significant attenuation,
3
hAM(22-52) blocked this response par- ranged between 4.3 Ϯ 0.2 and 4.6 Ϯ 0.3 found a significant population of
tially, to 110 Ϯ 3% and to 119 Ϯ 2%, pM and between 0.089 Ϯ 0.007 and 0.102 CGRP-R1 in intramural coronary arteries
respectively. Combined application of Ϯ 0.006 pM, respectively. LPS treatment of rats; however, they did not examine the
these drugs completely suppressed the significantly increased big ET-1 (approx- potential role of AM receptors. In the rat
AM-induced elevation of CF.
imately threefold) and ET-1 (twofold). In aorta, the existence and functional rele-
1
030
Crit Care Med 2001 Vol. 29, No. 5

Table 1. Pulmonary venous concentrations of big endothelin (ET)-1, ET-1, and adrenomedullin (AM) measured during serial perfusion
Big ET-1 (pM) ET-1 (pM) AM (pM)
Time
(
mins)
Control
LPS
Control
LPS
Control
LPS
a
a
a
a
a
a
a
a
a
0
1–20
1–30
–10
4.3 Ϯ 0.2
4.6 Ϯ 0.3
4.5 Ϯ 0.2
12.6 Ϯ 0.8
13.3 Ϯ 0.7
13.1 Ϯ 0.6
0.092 Ϯ 0.005
0.102 Ϯ 0.006
0.089 Ϯ 0.007
0.172 Ϯ 0.015
0.192 Ϯ 0.017
0.206 Ϯ 0.015
10.2 Ϯ 0.7
9.6 Ϯ 0.8
9.9 Ϯ 0.4
48.5 Ϯ 2.7
50.2 Ϯ 3.5
51.2 Ϯ 3.1
1
2
LPS, lipopolysaccharide.
a
p Ͻ .05 vs. control.
vance of both receptors—the CGRP-R1 Although involvement of smooth muscle neutrophils (45) may tip the balance be-
and the AM receptor—have been demon- ET receptors in coronary responses also tween AM and ET-1 toward coronary con-
B
strated recently (26).
has been reported, relevant ET -mediated striction. Considering the remote effects
B
In the isolated rat heart, Szokodi et al. effects usually are observed at markedly of pulmonary mediators, our study un-
34) found inotropic effects at AM con- higher (i.e., low nanomolar) concentra- derlines the potential usefulness of ET-1
(
centrations as low as 30 pM. Moreover, tions (35, 36).
antagonists as a therapeutic tool in
these authors documented vasodilatory
For both ET-1 and AM, significant ef- ARDS, because ET-1 blockade potentiates
effects at AM concentrations as low as 300 fects on cardiac inotropy have been re- AM-induced coronary dilation. Until now,
pM, although they used a model with ported (37–39). However, we were able to there have been no clinical data regard-
marked coronary predilation due to lim- detect specific inotropic effects neither ing administration of ET antagonists in
ited perfusion rate. Considering these dif- for big ET-1/ET-1 nor for AM. Other ex- human ARDS, sepsis, or systemic inflam-
ferences in experimental setup, the data periments, based on isolated hearts, also matory response syndrome (46). Compel-
indicate effectiveness of AM at concentra- have failed to demonstrate those effects ling evidence, however, has been gath-
tions that are comparable to those mea- for ET-1 (40). With respect to AM, the ered from different animal models
sured in pulmonary effluent in our LPS findings are even more controversial be- indicating that blockade of ET receptors
A
group (~50 pM) and to those used for cause positive (38), negative (39), and in- or mixed ET antagonism (ET and ET )
A
B
exogenous application in isolated hearts different effects (41) on inotropy have been improves systemic and pulmonary hemo-
100 pM). disclosed. Thus, the issue of AM and inot- dynamics (46), decreases retention of leu-
Parallel to the up-regulation of AM, ropy remains to be completely unraveled.
(
kocytes in the lung (47), and alleviates
expression of prepro-ET-1 mRNA and
synthesis of big ET-1 and ET-1 are in- future therapeutic strategies, we infer
For the in vivo situation as well as for nitric oxide-mediated lung injury (48).
We conclude that in experimental
creased in lung tissue (11, 27). We were from our model that both the vasocon- ARDS, elevated pulmonary AM—by occu-
able to confirm these findings by measur- strictor system (big ET-1 and ET-1) and pation of coronary CGRP type-1 receptors
ing elevated concentrations of big ET-1 the counteracting mediator (AM) are up- and AM receptors—outweighs the coro-
and ET-1 in pulmonary venous effluent of regulated in pulmonary vasculature and nary vasoconstrictor impact of pulmo-
LPS-stimulated lungs. On inhibition of may achieve pathophysiologically rele- nary big ET-1 exerted via ET receptors
A
AM-induced coronary vasodilation, we vant concentrations in pulmonary out- after conversion to mature ET-1. The
unmasked a coronary vasoconstrictor ef- flow. It also seems important that endog- consequence is prevention of flow-related
fect that was significantly stronger than enous pulmonary AM potentially may deterioration of myocardial performance.
that observed in control preparations and neutralize the deleterious remote effects
that evoked significant contractile de- of endogenous pulmonary ETs. From our
REFERENCES
pression. Conversely, blockade of ET re- model, we cannot precisely predict the
A
1
. Seeger W, Wolf H, Stahler G, et al: Increased
pulmonary vascular resistance and perme-
ability due to arachidonate metabolism in
isolated rabbit lungs. Prostaglandins 1982;
ceptors or inhibition of ECE markedly magnitude of coronary changes under
augmented coronary vasodilation in com- distinct in vivo conditions. It recently has
parison with that observed in untreated been demonstrated that plasma concen-
LPS preparations. Because coronary flow trations of AM increase dramatically (10-
response in LPS preparations was com- to 40-fold) in patients with sepsis and
pletely reversed to preserial baseline val- systemic inflammatory response syn-
ues by use of the triple combination of ET drome, with values detected as high as
and AM blockers (Fig. 2B), we can rea- 50 –200 pM (42– 44). Moreover, AM
sonably argue that the documented ef- plasma concentrations correlate with sys-
fects primarily reflect the opposite ac- temic and pulmonary vascular resistance
tions of elevated endogenous pulmonary (43) as well as with circulating concen-
2
3:157–172
2
. Sprague RS, Thiemermann C, Vane JR: En-
dogenous endothelium-derived relaxing fac-
tor opposes hypoxic pulmonary vasoconstric-
tion and supports blood flow to hypoxic
alveoli in anesthetized rabbits. Proc Nat Acad
Sci USA 1992; 89:8711–8715
3
4
. Perreault T, Gutkowska J: Role of atrial na-
triuretic factor in lung physiology and pa-
thology. Am J Respir Crit Care Med 1995;
AM and pulmonary big ET-1.
trations of tumor necrosis factor-␣ (44)
The effectiveness of BQ-123 to com- in these patients. These facts may accen-
pletely prevent coronary response to ex- tuate the potential clinical relevance of
ogenous big ET-1 and ET-1 as well as our findings. On the other hand, several
during serial perfusion emphasizes the factors including coronary endothelial
1
51:226–242
. Gerbes AL, Dagnino L, Nguyen T, et al: Tran-
scription of brain natriuretic peptide and
atrial natriuretic peptide genes in human
tissues. J Clin Endocrinol Metab 1994; 78:
1307–1311
prevailing contribution of ET receptors dysfunction (8, 9) and enhancement of
A
to ET effects in the low picomolar range. pulmonary release of ETs by activated
Crit Care Med 2001 Vol. 29, No. 5
1031

5
6
. Klinger JR, Siddiq FM, Swift RA, et al: C-type
natriuretic peptide expression and pulmo-
nary vasodilation in hypoxia-adapted rats.
Am J Physiol 1998; 275:L645–L652
. Dupuis J, Stewart DJ, Cernacek P, et al: Hu-
man pulmonary circulation is an important
site for both clearance and production
of endothelin-1. Circulation 1996; 94:
Attenuation of endotoxin-induced pulmo-
nary injury by antithrombin III. Am J Physiol
1996; 270:L921–L930
thelin receptor (ET and ET ) mediate vaso-
A B
constriction in the perfused rat heart.
Cardiovasc Pharmacol 1993; 22:
J
20. Ihara M, Noguchi K, Saeki T, et al: Biological
profiles of highly potent novel endothelin
antagonists selective for the ETA receptor.
Life Sci 1992; 50:247–255
S248–S251
36. Pernow J, Modin A: Endothelial regulation of
coronary vascular tone in vitro: Contribution
of endothelin receptor subtypes and nitric
oxide. Eur J Pharmacol 1993; 243:281–286
21. Von Geldern TW, Tasker AS, Sorensen BK, et
1
578–1584
al: Pyrrolidine-3-carboxylic acids as endothe- 37. Takanashi M, Endoh M: Characterization of
7
. Martinez A, Miller MJ, Unsworth EJ, et al:
Expression of adrenomedullin in normal
human lung and in pulmonary tumors.
Endocrinology 1995; 136:4099–4105
. Stangl K, Dschietzig T, Laule M, et al: Pul-
monary big endothelin affects coronary tone
lin antagonists. 4. Side chain conformational
restriction leads to ET(B) selectivity. J Med
Chem 1999; 42:3668–3678
positive inotropic effect of endothelin on
mammalian ventricular myocardium. Am J
Physiol 1991; 261:H611–H619
8. Szokodi I, Kinnunen P, Tavi P, et al: Evi-
dence for cAMP-independent mechanisms
mediating the effects of adrenomedullin, a
new inotropic peptide. Circulation 1998; 97:
22. Emoto N, Yanagisawa M: Endothelin-
converting enzyme-2 is a membrane-bound,
phosphoramidon-sensitive metalloprotease
with acidic pH-optimum. J Biol Chem 1995;
270:15262–15268
23. Sheykhzade M, Nyborg NC: Characterization
of calcitonin gene-related peptide receptors
in intramural coronary arteries from male
and female Sprague Dawley rats. Br J
Pharmacol 1998; 123:1464–1470
3
8
and leads to an enhanced, ET -mediated cor-
A
onary constriction in early endothelial dys-
function. Circulation 1997; 96:3192–3200
. Dschietzig T, Laule M, Alexiou K, et al: Cor-
onary constriction and consequent cardiode-
pression in pulmonary embolism are medi-
ated by pulmonary big endothelin and
enhanced in early endothelial dysfunction.
Crit Care Med 1998; 26:510–517
1
062–1070
3
9. Ikenouchi H, Kangawa K, Matsuo H, et al:
Negative inotropic effect of adrenomedullin
in isolated adult rabbit cardiac ventricular
myocytes. Circulation 1997; 95:2318–2324
0. Karwatowska-Prokopczuk E, Wennalm A: Ef-
fects of endothelin on coronary flow, me-
chanical performance, oxygen uptake and
formation of purines and on outflow of pros-
tacyclin in the isolated rabbit heart. Circ Res
9
4
24. Osajima A, Uezono Y, Tamura M, et al: Ad-
renomedullin-sensitive receptors are prefer-
entially expressed in cultured rat mesangial
cells. Eur J Pharmacol 1996; 315:319–325
1
0. Stangl K, Dschietzig T, Richter C, et al: Pul-
monary release and coronary and peripheral
consumption of big endothelin and endothe- 25. Gardiner SM, Kemp PA, March JE, et al:
lin-1 in severe heart failure: Acute effects of
vasodilator therapy. Circulation 2000; 102:
1
990; 66:46–54
Regional haemodynamic effects of human
and rat adrenomedullin in conscious rats.
Br J Pharmacol 1995; 114:584–591
4
4
1. Saetrum Opgaard O, de Vries R, Tom B, et al:
Positive inotropy of calcitonin gene-related
peptide and amylin on porcine isolated myo-
cardium. Eur J Pharmacol 1999; 385:
1
132–1138
1
1
1. Curzen NP, Kaddoura S, Griffiths MJD, et al:
Endothelin-1 in rat endotoxemia: mRNA ex-
pression and vasoreactivity in pulmonary and
systemic circulations. Am J Physiol 1997;
26. Yoshimoto R, Mitsiu-Saito M, Ozaki H, et al:
Effects of adrenomedullin and calcitonin
gene-related peptide on contractions of the
rat aorta and porcine coronary artery. Br J
Pharmacol 1998; 123:1645–1654
1
47–154
2. Hirata Y, Mitaka C, Sato K, et al: Increased
circulating adrenomedullin, a novel vasodi-
latory peptide, in sepsis. J Clin Endocrinol
Metab 1996; 81:1449–1453
43. Nishio K, Akai Y, Murao Y, et al: Increased
plasma concentrations of adrenomedullin
correlate with relaxation of vascular tone in
patients with septic shock. Crit Care Med
4
1:L921–L930
2. Shoji H, Minamino N, Kangawa K, et al:
Endotoxin markedly elevates plasma concen-
tration and gene transcription of ad-
renomedullin in rat. Biochem Biophys Res
Commun 1995; 215:531–537
3. Snapper JR, Thabes JS, Lefferts PL, et al:
Role of endothelin in endotoxin-induced sus-
tained pulmonary hypertension in sheep.
Am J Respir Crit Care Med 1998; 157:81–88
4. Kitamura K, Kangawa K, Kawamoto M, et al:
27. Stangl K, Dschietzig T, Alexiou K, et al: An-
tithrombin increases pulmonary endothe-
2
ϩ
lins: Inhibition by heparin and Ca antago-
nism. Eur J Pharmacol 1999; 370:57–61
28. Brunner F: Dependence of endothelin-1 se-
2
ϩ
1
cretion on Ca . Biochem Pharmacol 1995;
49:1785–1791
1
997; 25:953–957
4
4
4
4. Ueda S, Nishio K, Minamino N, et al: In-
creased plasma levels of adrenomedullin in
patients with systemic inflammatory re-
sponse syndrome. Am J Respir Crit Care Med
29. Bortz J, Lienert GA, Boehnke K: Verteilungs-
freie Methoden in der Biostatistik. Berlin,
Springer Verlag, 1990
1
Adrenomedullin: A novel hypotensive peptide 30. Holm S: A simple sequentially rejective mul-
isolated from human pheochromocytoma.
Biochem Biophys Res Commun 1993; 192:
1
999; 160:132–136
tiple test procedure. Scand J Stat 1979;
6:65–70
5. Schmeck J, Janzen R, Munter K, et al: Endo-
thelin-1 and thromboxane A2 increase pul-
monary vascular resistance in granulocyte-
mediated lung injury. Crit Care Med 1998;
5
53–560
31. Isumi Y, Shoji H, Sugo S, et al: Regulation of
adrenomedullin production in rat endothe-
lial cells. Endocrinology 1998; 139:838–846
32. Sugo S, Minamino N, Shoji H, et al: Inter-
leukin-1, tumor necrosis factor and lipopoly-
saccharide additively stimulate production of
adrenomedullin in vascular smooth muscle
cells. Biochem Biophys Res Commun 1995;
207:25–32
1
1
5. Samson WK: Adrenomedullin and the con-
trol of fluid and electrolyte homeostasis.
Annu Rev Physiol 1999; 61:363–389
6. National Institutes of Health: Guide for the
Care and Use of Laboratory Animals (NIH
Publication No. 85–23). Bethesda, MD, Na-
tional Institutes of Health, 1996
2
6:1868–1874
6. Wanecek M, Weitzberg E, Rudehill A, et al:
The endothelin system in septic and endo-
toxin shock. Eur J Pharmacol 2000; 407:
1–15
47. Sato Y, Hogg JC, English D, et al: Endothe-
lin-1 changes polymorphonuclear leuko-
cytes’ deformability and CD11b expression
and promotes their retention in the lung.
Am J Respir Cell Mol Biol 2000; 23:404–410
1
7. Baydoun AR, Foale RD, Mann GE: Bacterial
endotoxin rapidly stimulates prolonged en- 33. Entzeroth M, Doods HN, Wieland HA, et al:
dothelium-dependent vasodilation in the rat
isolated perfused heart. Br J Pharmacol
Adrenomedullin mediates vasodilation via
CGRP1 receptors. Life Sci 1995; 56:
PL19–PL25
1
993; 109:987–991
1
1
8. Uhlig S, Brasch F, Wollin L, et al: Functional 34. Szokodi I, Kinnunen P, Ruskoaho H: Inotro- 48. Fujii Y, Magder S, Cernacek P, et al: Endo-
and fine structural changes in isolated rat
lungs challenged with endotoxin ex vivo and
in vitro. Am J Pathol 1995; 146:1235–1247
9. Uchiba M, Okajima K, Murakami K, et al:
pic effect of adrenomedullin in the isolated
rat heart. Acta Physiol Scand 1996; 156:
151–152
thelin receptor blockade attenuates lipopo-
lysaccharide-induced pulmonary nitric oxide
production. Am J Respir Crit Care Med 2000;
161:982–989
35. Balwierczak JL: Two subtypes of the endo-
1
032
Crit Care Med 2001 Vol. 29, No. 5