OUP user menu

Mechanisms of elevated plasma endothelin-1 in CHF: congestion increases pulmonary synthesis and secretion of endothelin-1

Thomas G von Lueder, Harald Kjekshus, Thor Edvardsen, Erik Øie, Stig Urheim, Leif Erik Vinge, Muhammed Shakil Ahmed, Otto A Smiseth, Håvard Attramadal
DOI: http://dx.doi.org/10.1016/j.cardiores.2004.03.016 41-50 First published online: 1 July 2004

Abstract

Objective: The pulmonary circulation may contribute to elevated plasma levels of endothelin-1 (ET-1) in congestive heart failure (CHF). The aims of the present study were to determine the mechanisms of increased secretion of ET-1 from the pulmonary circulation in CHF. Methods: Juvenile pigs were subjected to sham operation (n=9) or rapid cardiac pacing-induced CHF (215–240 bpm, n=15). Results: Three weeks of rapid pacing led to significant left ventricular dilatation, increased cardiac filling pressures, and reduced contractility (CHF pigs). Arterial plasma ET-1 levels in the CHF pigs were increased 4-fold compared to sham pigs (P<0.001). Single-bolus multiple indicator-dilution experiments revealed that pulmonary synthesis and release of ET-1 was increased in CHF, while pulmonary clearance of plasma ET-1 remained unaltered despite significant reduction of pulmonary fractional extraction of plasma ET-1. Pulmonary ECE-1 isozyme activity (pmol·min−1·mg protein−1) was selectively increased in lower lobe segments of CHF pigs (2.0±0.3) compared to lower lobe segments of controls (1.1±0.1, P<0.02), and to upper lobe segments of CHF pigs (1.1±0.1, P<0.005), and correlated significantly with the wet/dry weight ratios of the pulmonary tissue samples (R=0.75, P<0.001), i.e. a marker of pulmonary congestion. Furthermore, alveolar macrophages in congested lobe segments were identified as likely sites of increased synthesis and release of ET-1. Conclusions: In rapid pacing-induced CHF, a complex cardiopulmonary interaction revealed by pulmonary congestion causes increased pulmonary production and secretion of ET-1 due to enhanced pulmonary ECE-1 activities. Pulmonary secretion of ET-1 during evolving CHF is an important contributor to elevated plasma ET-1 levels in the systemic circulation.

Keywords
  • Heart failure
  • Endothelin
  • Hemodynamics
  • Pulmonary circulation

1. Introduction

Substantial evidence from several laboratories points to an important role of the potent vasoconstrictor endothelin-1 (ET-1) in the pathogenesis and progression of congestive heart failure (CHF) [1]. Plasma ET-1 levels are elevated in CHF and correlate with clinical severity [2–4]. In addition, plasma ET-1 levels have been found to be strong predictors of mortality in CHF [1,5]. Previous studies, including a report from our laboratory, have shown that the pulmonary circulation contributes to circulating plasma ET-1 levels in CHF [6,7]. However, the mechanisms of increased pulmonary secretion of ET-1 remain obscure. Besides being a major site of synthesis of ET-1, the lungs also take part in removal of ET-1 from plasma [8]. In isolated perfused lungs of rats with CHF secondary to myocardial infarction, pulmonary extraction (percent of bolus) of trace amounts of [125I]-ET-1 was decreased [6]. However, little information on pulmonary clearance of ET-1 in patients with CHF versus healthy controls is available. In a study of patients with pulmonary hypertension, which also included a few patients with CHF, pulmonary clearance of ET-1 was reduced compared to a historic control group [9]. In a recent preliminary report pulmonary clearance of ET-1 in CHF patients was also reduced compared to a historic control group [10]. Thus, although previous data indicate that pulmonary clearance of ET-1 may be reduced in CHF, the relative contributions of pulmonary clearance versus pulmonary synthesis of ET-1 to elevation of plasma ET-1 levels in CHF are unknown.

The biosynthesis of ET-1 occurs through several proteolytic steps from inactive precursor peptides to biologically active peptide. The 203 amino acid translation product prepro-endothelin-1 is cleaved by furin-like proteases to form the inactive intermediate big endothelin-1 (big ET-1). Big ET-1 is subsequently processed by endothelin-converting enzymes (ECE) into biologically active ET-1. Two ECE isoforms have been molecularly cloned and characterized, ECE-1 and ECE-2, which efficiently convert big ET-1 to ET-1 albeit at very different pH optima [11,12]. In CHF, synthesis of active ET-1 may be regulated at the level of transcription of preproET-1 mRNA as well as post-translationally at the level of ECE activity. Current evidence indicates that ECE activity is the rate-limiting step in the synthesis of ET-1 in vivo and may be increased in CHF [13,14]. However, whether alterations of ECE activities lead to increased pulmonary synthesis and secretion of ET-1 has not been studied.

Therefore, the aim of the present study was to determine the mechanisms of increased secretion of ET-1 from the pulmonary circulation in CHF. In this respect, experiments were designed to quantify and differentiate between changes in pulmonary clearance versus synthesis of ET-1. Pulmonary ECE-1 and ECE-2 activities as well as ET-1 content were determined. To investigate the putative influence of pulmonary congestion secondary to CHF, ECE activities were determined in samples from the upper lobes versus the dependent lower lobes, and related to pulmonary wet weight/dry weight ratios. Finally, to identify the cellular sites of ET-1 synthesis, immunohistochemical analysis of pulmonary tissue sections was performed.

2. Methods

2.1. Animal model

The investigation conforms with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996). Juvenile Norwegian farm pigs (average body weight 35 kg) of either sex were subjected to rapid pacing-induced heart failure (CHF pigs, n=15) or sham-operation (sham pigs, n=9) as described previously [7]. Briefly, in the CHF group a transvenous electrode was positioned in the right ventricle and connected to a programmable pacemaker which was buried in a subcutaneous pocket. Cefalotin (1 g im.) and buprenorphine (0.3 mg im. bid) were given to all pigs the first postoperative day. After recovery for 5–7 days, CHF was induced by rapid pacing (200–240 bpm, 21 days) as described previously [7].

2.2. Echocardiography, hemodynamic measurements, and blood sampling

At day 21 rapid pacing was terminated. Thirty minutes after cessation of pacing, general anesthesia was induced and echocardiography and hemodynamic measurements were performed as described previously [7]. Left ventricular fractional shortening (FS) and ejection fraction (EF) were obtained from short-axis views, using two-dimensional targeted M-mode recordings at the level of the papillary muscles. The recordings were analysed by two trained specialists.

Hemodynamic measurements were performed using a 5F micromanometer-tipped catheter (Millar Instruments) in the left ventricle, and a 7F Swan-Ganz thermodilution catheter (Baxter) in the pulmonary artery. Collection catheters for venous blood sampling were introduced via the jugular veins and advanced under fluoroscopic guidance to ascertain correct anatomical position. Arterial and venous blood samples were drawn simultaneously into prechilled vacutainers containing 1 mg/ml EDTA. Plasma was recovered and stored at −70 °C.

For time course analysis of ET-1 release from the pulmonary circulation during evolving heart failure, 4 sham pigs and 7 CHF pigs were permanently instrumented with introducer sheaths to allow insertion of the beforementioned catheters and collection of blood samples at day 0, 1, 7, 14 and 21 after initiation of pacing.

2.3. Indicator-dilution experiments and analysis of pulmonary ET-1 metabolism

In order to estimate pulmonary clearance of plasma ET-1, the single-bolus multiple indicator-dilution technique was applied in 5 of the sham pigs and 8 of the CHF pigs [8]. Briefly, a 2 ml bolus containing trace amounts of [125I]-ET-1 and a non-metabolizable vascular reference (Evans' blue dye-labeled albumin) was injected into the pulmonary artery outflow tract, and timed sequential blood collection was simultaneously initiated from a collection catheter positioned in the left ventricle. Evans' blue-labeled albumin and [125I]-ET-1 in the outflow samples were measured as previously described [15]. Fractional recoveries per ml blood for Evans' blue-labeled albumin and [125I]-ET-1 as function of time were plotted in a logarithmic ordinate format to generate indicator-dilution curves. The areas under the extrapolated curves for [125I]-ET-1 relative to that of Evans' blue-labeled albumin provided an estimate of fractional pulmonary survival of ET-1, from which mean pulmonary extraction of ET-1 could be calculated. Based on the values for pulmonary ET-1 extraction (EXT), hematocrite (hct), cardiac index (CI), plasma ET-1 levels in the pulmonary artery (ET-1PA), and the plasma ET-1 gradient over the lungs (ET-1grad), pulmonary clearance, synthesis and survival of ET-1 was calculated as previously published [8], according to the following formulae:

  1. Pulmonary ET-1 clearance=[(1−hct)×CI]×ET-1PA×EXT

  2. Pulmonary ET-1 synthesis=[(1−hct)×CI]×ET-1grad+Pulmonary ET-1 clearance

  3. Pulmonary ET-1 survival=[(1−hct)×CI]×ET-1PA×(1−EXT).

2.4. Tissue sampling

After completion of hemodynamic recordings and blood sampling, the pigs were euthanized by injection of pentobarbital (30 mg/kg, i.v.). Tissue samples were obtained from pulmonary upper-lobe (cephalad-dorsal) and lower-lobe (caudal–ventral) segments, the main proximal and distal pulmonary artery, and the main pulmonary veins. Tissue samples for immunohistochemistry were obtained from three CHF pigs and two sham pigs, rinsed in saline, and subsequently fixed in Bouin's solution (2% paraformaldehyde and 0.2% picric acid in phosphate buffered saline).

2.5. Plasma and tissue levels of ET-1

Plasma ET-1 levels were determined by direct chemiluminescent ELISA (Quantiglo QET 00, R&D Systems, USA). Cross-reactivities with other ET-peptides were as follows: human ET-2, 27.4%; human ET-3, 7.8%; porcine big ET-1, 0.02%.

Samples for determination of tissue ET-1 levels were processed according to an established protocol for extraction of ET-1 prior to colorimetric ELISA of ET-1 (RPN228, Amersham Biosciences) [16]. This ELISA system exhibited cross-reactivity with other ET peptides as follows: ET-2, >100%; ET-3, <0.001%; porcine big ET-1, ∼0.002%. The recovery of the extraction procedure was 72±2% as determined by addition of various amounts of exogenous ET-1 to the samples.

2.6. Assay of ECE activities

Solubilized membrane suspensions were prepared as described previously [11]. Assay of ECE-1 activities in the solubilized membranes was carried out at 37 °C for 30 min in 100 mmol/l sodium phosphate buffer [pH 6.8] containing 0.6 μmol/l porcine big ET-1 [1–38] and the presence or absence of 0.6 mmol/l of the ECE-selective inhibitor FR901533 (a kind gift from Fujisawa, Japan) to determine specific cleavage of big ET-1 [12]. Assay of ECE-2 activities was carried out as described above, except that the reaction buffer used was 100 mmol/l MES–HCl [pH 5.5] [12]. The assay was stopped by adding EDTA to 2.5 mmol/l, and the amount of ET-1 formed was determined by ELISA (RPN228, Amersham Biosciences). In our experiments the cross-reactivity with big ET-1 was negligible at the concentrations of big ET-1 employed. Inter- and intra-assay variations were <10% and 6–8%, respectively.

2.7. Immunohistochemistry

Fixed tissue samples embedded in paraffin wax were cut in 5 μm sections, processed, and subjected to immunohistochemical analysis as previously described [17]. The antibodies used were rabbit anti-pig preproET-1 antiserum (Peninsula Laboratories, USA; dilution of 1:500), and rabbit anti-human ECE-1 antiserum (a generous gift from Dr. Florence Pinet, INSERM U36, Paris; dilution 1:300) [18]. Antigenic epitopes were unmasked by heating the sections in sodium citrate buffer, pH 6.0, in a microwave oven. The immunoreactivities were amplified by the avidin–biotin–peroxidase system (Vector Laboratories, USA). Diaminobenzidine was used as the chromogen in a commercial metal enhanced system (Pierce Chemical, USA). The sections were counterstained with hematoxylin. Nonimmune normal rabbit serum or omission of primary antibody was employed as negative controls.

2.8. Statistical analysis

All values are reported as mean±S.E.M. Paired or unpaired Student's t-test was used as appropriate to assess differences between groups, and linear or nonlinear regression was used to test the relationship between the variables. Normal distribution of variances was confirmed by Mann–Whitney test. P<0.05 was considered significant.

3. Results

3.1. Echocardiography

Echocardiographic analysis revealed significant cardiac dilatation in the pacing group, as evidenced by increased left ventricular end-diastolic diameters and septal thinning (Table 1). The CHF pigs also showed significantly reduced contractility, as revealed by decreased fractional shortening and ejection fraction.

View this table:
Table 1

Study parameters of sham and CHF group

Sham (n=9)CHF (n=15)
LV end-diastolic diameter, cm/m24.5±0.25.6±0.2*
Interventricular septum sys, cm/m21.2±0.10.8±0.1*
Fractional shortening, %38±212±2
Ejection fraction, %76.0±2.631.7±3.6
Heart rate, bpm108±9103±5
MAP, mm Hg85±663±4*
PAP systolic, mm Hg21.4±1.839.0±1.6*
PAP mean, mm Hg15.5±1.830.3±1.3*
LVEDP, mm Hg9.6±1.928.5±1.3*
PCWP, mm Hg7.0±1.618.2±1.1*
CVP, mm Hg3.7±1.312.2±0.8*
SVR mm Hg/l/min/m210.8±0.715.8±1.5*
PVR mm Hg/l/min/m21.4±0.35.0±1.1*
Cardiac Index, l/min/m27.9±0.64.2±0.2*
dP/dtmax, mm Hg/s1711±115788±67*
dP/dtmin, mm Hg/s−1907±147−818±74*
Tau (τ), ms41.9±4.468.3±3.2
Total heart weight, g/m2230±10341±12
Left ventricular weight, g/m2128±6153±7
Right ventricular weight, g/m243.8±2.664.9±3.6
Atrial weight, g/m239.0±2.086.4±4.0
Liver weight, g/m21035±151822±99
Lung weight, g/m2497±20746±47
Wet/dry weight ratio UL5.3±0.15.4±0.1
Wet/dry weight ratio LL5.2±0.15.8±0.1§
  • Values are means±S.E.M. LV, left ventricle; MAP, mean arterial pressure; PAP, pulmonary artery pressure; LVEDP, left ventricular end-diastolic pressure; PCWP, pulmonary capillary wedge pressure; CVP, central venous pressure; SVR, systemic vascular resistance; PVR, pulmonary vascular resistance; UL, upper lung lobes; LL, lower lung lobes.

  • * P<0.01, CHF compared to sham pigs.

  • P<0.001, CHF compared to sham pigs.

  • P<0.02, CHF compared to sham pigs.

  • § P<0.02, LL CHF vs. UL CHF.

3.2. Hemodynamic measurements

The results of the invasive hemodynamic measurements are shown in Table 1. The CHF pigs exhibited significant increases in both left and right ventricular filling pressures. Cardiac index was reduced by 50% (P<0.001), and pulmonary and systemic vascular resistances were 3.5 times (P<0.001) and 1.5 times (P<0.05), respectively, that of sham pigs. Impaired systolic and diastolic function in the CHF pigs was confirmed by decreased maximal and minimal LV dP/dt and increased τ (time constant of isovolumetric relaxation; P<0.001).

3.3. Organ weights

Cardiac hypertrophy was evidenced by increased cardiac chamber weights in the CHF pigs versus sham pigs (Table 1). Total lung and liver weights were also substantially increased in the CHF pigs (P<0.001). Furthermore, pulmonary tissue samples from the upper and lower lobes were weighed immediately after excision, and after vacuum drying. The wet/dry weight ratios were significantly elevated in the lower lobes versus the upper lobes of the CHF pigs as well as compared with parallel samples from sham pigs, indicating pulmonary congestion (Table 1).

3.4. Plasma ET-1 levels and AV-gradients

Plasma ET-1 levels in all regions sampled were substantially elevated in CHF pigs after 21 days of rapid cardiac pacing as compared to time-matched sham-operated pigs (Fig. 1A). No differences of plasma ET-1 levels were found in the left ventricle versus the pulmonary artery in sham pigs (Δ-0.02±0.04 pmol/l, ns.). In the CHF pigs, on the other hand, a substantial pulmonary arterio-venous gradient of plasma ET-1 levels was found (Δ0.67±0.13 pmol/l, P<0.001), demonstrating net secretion of ET-1 in the pulmonary circulation. On the contrary, net extraction of plasma ET-1 was observed in the transhepatic portal circulation of CHF pigs (portal vein 3.21±0.42 vs. hepatic vein 2.06±0.37 pmol/l; P<0.001). In a subset of five CHF pigs, blood was sampled from the iliac vein and from the caval vein immediately proximal to the renal vein. Plasma ET-1 levels in blood sampled from this location of the inferior caval vein were significantly lower than plasma ET-1 levels in the iliac veins (2.57±0.64 vs. 3.10±0.71 pmol/l; P<0.05), as well as those in aortic blood samples (3.31±0.37 pmol/l; P<0.05), indicating net extraction of plasma ET-1 in the renal circulation. Thus, the pulmonary circulation was the only major vascular bed that made a positive contribution to plasma ET-1 in CHF.

Fig. 1

(A) Regional plasma ET-1 levels in CHF pigs (closed bars) after 21 days of rapid cardiac pacing versus time-matched sham pigs (open bars). The upper line indicates mean aortic ET-1 level in CHF pigs, whereas the dashed lower line indicates mean aortic ET-1 level in sham pigs. *P<0.001 compared to sham pigs; †P<0.05 versus mean aortic values in CHF pigs; #P<0.01 versus mean aortic values in sham pigs. (B) Time course of pulmonary arterio-venous gradient of plasma ET-1 in sham pigs (○) and in CHF pigs (●). The AV-gradients were determined as the difference of plasma ET-1 levels in the left ventricle and those in the pulmonary artery at day 0 (baseline), 1, 7, 14, and 21 after initiation of rapid pacing or sham operation. Values are mean±S.E.M. *P<0.05 versus baseline (ANOVA); †P<0.05 versus time-matched sham pigs.

Repeated blood sampling during cardiac pacing allowed us to determine the time course of ET-1 secretion in the pulmonary circulation. As shown in Fig. 1B, the plasma ET-1 levels in the left ventricle and those in the pulmonary artery demonstrated increasing arterio-venous differences over the pulmonary circulation during evolving pacing-induced heart failure.

3.5. Pulmonary clearance and synthesis of ET-1

Pulmonary indicator-dilution experiments for [125I]-ET-1 and Evans' blue dye-labeled albumin were performed as described in Methods. Fig. 2A shows representative pulmonary outflow indicator-dilution curves from a sham pig and a CHF pig plotted in logarithmic ordinate format.

Fig. 2

(A) Representative set of pulmonary outflow indicator-dilution curves in sham and CHF pigs. Fractional recoveries per ml blood as a function of time for the vascular reference indicator Evans' blue dye-labeled albumin (○) and [125I]-ET-1 (●) are presented in logarithmic ordinate formats. Linear extrapolation of the semilogarithmic downslopes (solid and dashed lines) of the curves is plotted to define the first passage of the two indicators. (B) Natural logarithm of the ratio of the fractional recoveries per ml blood for Evans' blue dye-labeled albumin and [125I]-ET-1 as a function of time in sham pigs (□) and CHF pigs (■). The plots are from the same data set as in (A). Linear regression lines are indicated.

As shown in Fig. 2B, a plot of the natural logarithm of the ratio of the fractional recovery of Evans' blue albumin relative to that of [125I]-ET-1 increased linearly as function of time over the entire primary dilution period both in sham pigs and CHF pigs. Thus, extraction of ET-1 by the pulmonary circulation appeared to be a unidirectional process without any significant return of extracted tracer to the circulation during pulmonary transit in any of the groups. The uptake or sequestration rate constant (Kseq) characterizing pulmonary extraction of ET-1 was estimated from the slope of the curves, as described previously [15]. In CHF pigs, the mean rate constant Kseq was significantly lower than in sham pigs (0.0076±0.0020 s−1 vs. 0.0959±0.0168 s−1; P<0.0001), indicating that the kinetics of pulmonary extraction of ET-1 were substantially reduced in the CHF pigs. Consistently, pulmonary extraction of ET-1 (% removal of ET-1 during single pulmonary transit) in the CHF group was reduced to approximately 50% of that in sham pigs (P<0.02, Fig. 3B). However, pulmonary clearance of ET-1, i.e. the absolute amount of ET-1 cleared from plasma per minute, did not display significant differences in CHF versus sham pigs. On the contrary, pulmonary synthesis of ET-1 and survival of ET-1 were significantly increased in the CHF pigs (Fig. 3B). Furthermore, pulmonary synthesis of ET-1 displayed robust correlation with LVEDP and PCWP (Figs. 4A and B).

Fig. 4

Scatter plot of pulmonary synthesis of ET-1 in CHF pigs (●) at day 21 after initiation of pacing and in time-matched sham pigs (○) versus left ventricular end-diastolic pressure (LVEDP, A) and pulmonary capillary wedge pressure (PCWP, B). Linear regression (solid line) and 95% confidence interval (dashed lines) are indicated.

Fig. 3

(A) Mean pulmonary extraction of [125I]-ET-1 (% of vascular reference) during single pulmonary transit in sham pigs (open bar; n=5) versus CHF pigs (closed bar; n=8), calculated as described in Methods. *P<0.02 versus sham pigs. B. Quantitative determination of pulmonary clearance, synthesis and survival of ET-1 in sham pigs (open bars; n=5) versus CHF pigs (closed bars; n=8), computed as described in Methods. *P<0.01 versus sham pigs.

3.6. Pulmonary ECE activities and tissue ET-1 levels

As shown in Fig. 5A, ECE-1 activities were highest in congested lower lobes of CHF pigs compared to lower lobes of sham pigs (2.0±0.3 vs. 1.1±0.1 pmol·min−1·mg protein−1; P<0.02), and to upper lobes of CHF pigs (1.1±0.1 pmol·min−1·mg protein−1; P<0.005). Furthermore, the ECE-1 activities correlated significantly with pulmonary wet/dry weight ratios (Fig. 5C). ECE-2 activities, however, were not significantly different in the upper and lower lobes, nor between the two groups (Fig. 5B). The ECE-inhibitor FR901553 inhibited approximately 92% and 76% of total ECE-activities under the assay conditions optimal for ECE-1 and ECE-2, respectively. Furthermore, ECE-1 activities in the proximal and distal pulmonary artery and vein were very low compared to the activities in pulmonary tissue, and no differences of samples from the sham and CHF groups were found in these tissues (data not shown). Pulmonary ET-1 levels were increased 4- and 5-fold in the upper and lower lobes of CHF animals, respectively, as compared to controls (P<0.01). However, the pulmonary ET-1 levels in the upper versus the lower lobes of CHF animals were not statistically different (Fig. 6).

Fig. 6

Pulmonary tissue ET-1 levels in samples from upper lobes (UL) and lower lobes (LL) of sham pigs (open bars; n=5) and CHF pigs (closed bars; n=6). Tissue levels of ET-1 were determined as described in Methods. Data presented as mean±S.E.M. *P<0.05 versus sham pigs.

Fig. 5

Assay of pulmonary ECE activities in CHF pigs and in sham-operated pigs. (A) ECE-1 activities in tissue samples from upper lobes (UL) and lower lobes (LL) shown as matched pairs of individual sham pigs (○) and CHF pigs (●). Values are mean±S.E.M. *P<0.02 versus LL of sham pigs; †P<0.01 versus UL of CHF pigs. (B) ECE-2 activities in tissue samples from upper lobes (UL) and lower lobes (LL) of sham pigs (open bars, n=5) and CHF (closed bars, n=6) pigs. Values are mean±S.E.M. No significant differences of ECE-2 activities were found. (C) Scatter plot demonstrating ECE-1 activities in pulmonary tissue samples (data presented in (A)) of sham pigs (○) and CHF pigs (●) versus pulmonary wet/dry weight ratios. Linear regression (solid line) and 95% confidence interval (dashed lines) are indicated.

3.7. Immunohistochemical analysis of pulmonary tissue

Immunohistochemical analysis revealed similar distribution of immunoreactive ppET-1 and ECE-1 in pulmonary tissue sections (Fig. 7, Panels B and C), compared to non-immune staining (A). Anti-ppET-1 and anti-ECE-1 immunostaining were observed in bronchial and alveolar epithelial cells, as well as in vascular endothelial and smooth muscle cells. In alveolar tissue of the upper pulmonary lobes of both sham pigs (E) and CHF pigs (F), robust immunostaining of alveolar epithelial cells and vascular endothelial cells was observed. In CHF pigs (Panel F, upper lobe; Panel H, lower lobe), histochemical analysis of pulmonary tissue sections exhibited overt evidence of congestion with atelectasis and alveolar interstitial thickening. Pulmonary macrophages, which were rare in pulmonary tissue of sham pigs, also displayed intense anti-ppET-1 and anti-ECE-1 immunoreactivities. In the CHF pigs the abundance of immunoreactive macrophages was increased in congested lower lobe segments.

Fig. 7

Histology and distribution of anti-ECE-1 and anti-preproET-1 immunoreactivities in pulmonary tissue sections of sham pigs and CHF pigs. Panels A–B: Photomicrographs (magnification 200×) of serial sections of pulmonary tissue from sham pigs immunostained with non-immune serum (Panel A) or anti-ECE-1 antiserum (Panel B). Bronchial and alveolar epithelial cells, as well as vascular endothelial and smooth muscle cells demonstrate robust immunostaining. Panel C: Pulmonary section of sham pig immunostained with anti-preproET-1 antiserum (magnification 200×) Note: Similar distribution of anti-ECE-1 and anti-preproET-1 immunoreactivities can be seen (Panel B vs. Panel C). Panel D: Photomicrograph of section from the lower pulmonary lobe of CHF pig immunostained with anti-ECE-1 immunserum (magnification 400×). Panels E–H: High power magnification (400×) of distribution of anti-ECE-1 immunoreactivity in pulmonary sections of sham (Panels E and G) versus CHF pigs (Panels F and H). Similar distribution of anti-ECE-1 immunoreactivity can be observed in pulmonary sections from upper lobe (Panel E) and lower lobe (Panel G) of sham pigs. In CHF, pulmonary sections of upper lobe (Panel F) demonstrate interstitial thickening of alveolar walls and pronounced infiltration of immunoreactive macrophages. Histological architecture of lower lobe (Panel H) is found even more deranged with atelectasis, interstitial thickening, and massive infiltration of immunoreactive macrophages.

4. Discussion

The present study provides new knowledge of the pathophysiologic mechanisms of elevated plasma ET-1 in CHF. We demonstrate that pulmonary secretion of ET-1 increases with progression of heart failure and provides major contribution to elevation of plasma ET-1 levels. Our data provide compelling evidence that the mechanism of increased release of ET-1 from the lungs is increased pulmonary synthesis of ET-1. We report for the first time that the activity of ECE-1, a key enzyme in the biosynthesis of ET-1, is specifically increased in pulmonary tissue segments affected by pulmonary congestion. Furthermore, we provide evidence that increased pulmonary ECE-1 activities and ET-1 synthesis are related to congestion of alveolar tissue rather than increased pressures in the pulmonary artery and vein.

In the present study of pacing-induced heart failure, net extraction of plasma ET-1 was observed across the portal hepatic and renal circulation. Furthermore, plasma ET-1 levels in the coronary sinus and in the iliac veins were not significantly different from those in the aorta. Thus, neither the coronary circulation nor the lower extremities appear to contribute to elevated plasma ET-1 in pacing-induced heart failure in pigs. Ischemic myocardial tissue has been shown to express high levels of ET-1 [17]. Accordingly, in heart failure induced by myocardial infarction ischemic myocardial tissue could conceivably contribute to elevated plasma ET-1 by spillover in the coronary circulation. This notion, however, is not supported by data from studies of transcardiac arterio-venous gradients of plasma ET-1 in patients with ischemic heart failure [19,20] demonstrating transcardiac extraction of plasma ET-1. Similar net extraction of plasma ET-1 has also been reported in studies of arterio-venous gradients of the lower extremities in patients with CHF [19,21]. Thus, the pulmonary circulation stands out as a major contributor to elevated plasma ET-1 levels in CHF, although contributions from other organs cannot be completely excluded. Previous reports have indicated that the lungs may impact on plasma ET-1 levels both as sites of synthesis and secretion of ET-1 and as sites of removal of ET-1 from plasma [8,15]. Several groups have demonstrated decreased fractional extraction of ET-1 in the pulmonary circulation both in heart failure patients and in experimental models of CHF. However, quantitative analyses of pulmonary secretion versus clearance of ET-1 in health versus in heart failure have not been reported, and thus, their relative contributions to arterial plasma ET-1 levels in CHF have remained unknown. The present study demonstrates that the principal mechanism of increased pulmonary secretion of ET-1 in CHF is increased synthesis of ET-1. Although affliction of pulmonary tissue by increased pulmonary capillary pressure is conceivably similar in nature in heart failure of different etiologies, a few reports have failed to demonstrate differences in arterial and venous levels of plasma ET-1 across the pulmonary circulation in heart failure [10]. Such discrepancies could be accounted for by varying degrees of heart failure. Indeed, elevation of plasma ET-1 levels in heart failure is related to the severity of the disease according to NYHA classification [1]. In this respect, as demonstrated in the present study, pulmonary secretion of ET-1 increased during evolving heart failure.

The indicator-dilution experiments revealed unidirectional uptake of ET-1 both in CHF and sham pigs, albeit with dramatically lower rate constant (Kseq) in the CHF pigs. Decreased extraction of ET-1 from the pulmonary circulation in CHF has previously been attributed to decreased expression of endothelial ETB receptors [14,22]. Indeed, ETB receptor blockade was shown to essentially block pulmonary uptake of ET-1 in dogs by severely attenuating Kseq [23]. Despite these findings, however, pulmonary clearance of plasma ET-1 was not significantly different in the CHF pigs compared with sham. Consequently, the principal mechanism of elevated plasma ET-1 in the CHF pigs appears to be increased pulmonary synthesis and secretion of ET-1 to the pulmonary circulation. However, the precise biosynthetic steps implicated in increased pulmonary synthesis of ET-1 remained to be characterized. Interestingly, an illuminating finding of the present study was that pulmonary ECE-1 activities were substantially increased in the CHF pigs and reflected in increased pulmonary tissue contents of ET-1. The induction of ECE activities appeared to be restricted to the ECE-1 isozyme activity since no alterations of ECE-2 activity were found. Furthermore, ECE-1 activity constituted the predominant ECE activity assayed in pulmonary tissue as judged from assays performed at the pH optima of ECE-1 (pH 6.8) and ECE-2 (pH 5.5), respectively [11,12]. The ECE activity assayed was almost completely inhibited by the ECE-1/ECE-2 selective antagonist FR901553, excluding significant contributions from other metalloproteases such as NEP, which have been shown to cleave big ET-1 into mature ET-1. The most remarkable increase of ECE-1 activities was found in pulmonary tissue samples from congested lower lobes. Several additional findings point to the decisive role of pulmonary congestion as stimulus of increased ECE-1 activities. First, pulmonary ECE-1 activities correlated significantly with pulmonary wet weight/dry weight ratios, i.e. the degree of pulmonary congestion. Robust positive correlation between indices of cardiac filling pressures (PCWP and LVEDP) and pulmonary synthesis of ET-1 was also found. Secondly, immunohistochemical analysis revealed substantially increased anti-ECE-1 and anti-preproET-1 immunostaining of pulmonary sections from congested lower lobes of CHF pigs due in large part to increased abundance of alveolar macrophages. Thirdly, increased ECE-1 activities in CHF pigs did not appear to involve vascular ECE-1 since ECE-1 activities of the main pulmonary artery and vein were substantially lower than pulmonary ECE-1 activities and remained unaltered despite significant arterial and venous hypertension in the CHF pigs. However, smaller-sized vessels were not investigated and could conceivably also be sites of altered ECE-1 activities.

Immunohistochemical analysis of pulmonary tissue sections revealed that vascular endothelial cells, bronchial and alveolar epithelial cells, and macrophages were the principal sites of ET-1 biosynthesis, confirming earlier reports [24,25]. Increased abundance of immunoreactive alveolar macrophages in sections of congested lung also calls for a role of proinflammatory mechanisms in activation of pulmonary ECE-1. The most readily explicable proinflammatory mechanism would be local responses to transudation of plasma proteins in congested pulmonary segments leading to increased pulmonary influx of immunoreactive macrophages. Systemic proinflammatory mechanisms appear less likely, since vascular ECE-1 activities remained unaltered. Hypoxia may constitute another putative mechanism of increased synthesis of ET-1. Indeed, pulmonary tissue hypoxia in heart failure may be a relevant mechanism arising from hypoperfusion due to both reduced mean arterial pressure and pulmonary congestion. In a recent report, hypoxia-inducible factor-1 (HIF-1) was found to bind and activate the promoter of the human prepro-endothelin-1 gene [26]. However, to what extent HIF-1 binds and activates the promoter of the ECE-1 gene remains to be investigated.

In conclusion, the present study demonstrates the major role of congested pulmonary tissue as contributor to elevated plasma ET-1 levels in rapid pacing-induced CHF. The novel mechanisms of increased pulmonary production and secretion of ET-1 unraveled in the present study emphasize the importance of the complex pathophysiological interaction between the heart and the lungs in CHF. In view of the disappointing outcomes of early clinical trials of ET receptor antagonists in treatment of heart failure, our study provides rationale for developing alternative pharmacotherapeutical strategies that specifically target pulmonary synthesis and secretion of ET-1 in CHF. Endothelin-converting enzyme inhibition may be one such avenue currently receiving increasing attention [27] as putative addition to our pharmacotherapeutical armamentarium in heart failure.

Acknowledgments

This work was supported by Grants from the Norwegian Council on Cardiovascular Diseases and the Research Council of Norway. Birthe Mikkelsen is acknowledged for excellent technical assistance.

Footnotes

  • Time for primary review 18 days

References

  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
View Abstract