Journal Logo

Articles

Chronic Dual Inhibition of Angiotensin-Converting Enzyme and Neutral Endopeptidase During the Development of Left Ventricular Dysfunction in Dogs

Thomas, Chadwick V.; McDaniel, George M.; Holzgrefe, Henry H.*; Mukherjee, Rupak; Hird, R. Barry; Walker, Jennifer D.; Hebbar, Latha; Powell, James R.*; Spinale, Francis G.

Author Information
Journal of Cardiovascular Pharmacology: December 1998 - Volume 32 - Issue 6 - p 902-912
  • Free

Abstract

Atrial natriuretic peptide (ANP) is a 28-amino-acid peptide synthesized and stored in membrane-adherent granules in atrial cardiocytes and is released into the systemic circulation in response to atrial stretch (1,2). ANP binds and activates a receptor located in several organ systems including brain, kidney, heart, and systemic vasculature (3). After ANP-receptor activation, increased intracellular levels of guanosine 3′,5′-cyclic monophosphate (cGMP) occur, which results in a number of physiologic effects including natriuresis, decreased systemic vascular resistance, and inhibition of the renin-angiotensin-aldosterone system (1). In patients with chronic left ventricular (LV) dysfunction, serum levels of ANP are greatly increased and appear to correlate with the severity of symptoms (2,4). It has been hypothesized that initially, increased ANP improves LV-pump function by favorably altering loading conditions (1). However, these potential beneficial effects of ANP diminish with chronically increased circulating levels of ANP in chronic LV dysfunction (2). ANP is primarily degraded by a specific zinc metalloprotease, neutral endopeptidase (NEP) 24.11, which is located in the lung and proximal tubules of the kidney, as well as in other organ systems (5,6). In LV dysfunction, NEP inhibition with subsequent potentiation of serum ANP levels was shown to increase natriuresis and improve LV-pump function (7-12). For example, Margulies et al. (11) reported that acute administration of a NEP inhibitor in dogs with pacing-induced LV dysfunction increased renal cGMP production and natriuresis after exogenous ANP administration. However, in developing LV dysfunction, the effects of chronic NEP inhibition on indices of LV function remain to be defined.

Clinical studies clearly demonstrated that angiotensinconverting enzyme (ACE) inhibition provided beneficial effects on indices of LV function and prolonged survival in patients with chronic LV dysfunction (13-15). A past report demonstrated a relation between ACE activity and ANP-mediated cGMP levels (16). In rodent models of LV hypertrophy, a synergistic effect was observed with combined ACE and NEP inhibition with respect to modulation of LV mass (9,11)Moreover, in dogs with pacing-induced LV dysfunction, Seymour et al. (10) reported improved systemic hemodynamics after acute combined ACE and NEP inhibition (10). In dogs with pacing-induced LV dysfunction, this laboratory demonstrated that chronic ACE inhibition provided direct and beneficial effects on LV geometry and myocyte contractility (17). However, the specific effects of chronic combined ACE and NEP inhibition on LV and myocyte geometry and function in developing LV dysfunction remain unknown. Accordingly, the goal of our study was to use a rapid-pacing model of LV dysfunction to investigate the specific effects of long-term treatment with a dual-acting metalloprotease inhibitor (DMPI), which possesses both ACE- and NEP-inhibition activity, on LV and myocyte geometry and function.

METHODS

Model of pacing-induced LV dysfunction

Twenty-five adult mongrel dogs of either sex (age 9-16 months, 15-25 kg; Hazelton, Kalamazoo, MI, U.S.A.) were used in this study. The animals were instrumented serially to measure LV and arterial pressures as well as to obtain plasma samples. In addition, a pacemaker and stimulating electrode were implanted to produce rapid right ventricular pacing. The animals were induced with thiopental (2 mg/kg, Pentothal; Abbot Laboratories, Chicago, IL, U.S.A.), intubated, and ventilated with 100% oxygen. Maintaining a surgical plane of anesthesia with 1-3% isoflurane (Aurthan; Anaquest, Madison, WI, U.S.A.), a left thoracotomy was performed, and a shielded stimulating electrode was sutured onto the right ventricular outflow tract, connected to a programable pacemaker modified for programing heart rates ≤300 beats/min (Spectrax 5985; Medtronic, Inc., Minneapolis, MN, U.S.A.) and buried in a subcutaneous pocket. A previously calibrated microtipped transducer (model p5-X4; Konigsberg Instruments, Pasadena, CA, U.S.A.) was placed into the LV chamber through an small incision at the apex. The connection of the LV transducer was tunneled and externalized in the suprascapular region of each animal. The pericardium was left open, the incision closed, and the pleural space evacuated of air. Next, the right carotid artery was exposed and a vascular-access port (model GPV, 9F; Access Technologies, Skokie, IL, U.S.A.) was placed in the artery, advanced to the aortic arch, and sutured in place for subsequent arterial blood pressure measurements and blood sampling. The animals were allowed a 14-day recovery period at which time proper operation of all implanted instrumentation was confirmed. All animals used in this study were treated and cared for in accordance with the National Institutes of Health "Guide for the Care and Use of Laboratory Animals" (National Research Council, Washington, DC, 1996).

Experimental design

After recovery from the surgical procedure, baseline LV pressure and dimensions and arterial pressure were measured, and plasma samples were obtained for each dog, as described in the following sections. The pacemakers were then activated for rapid ventricular pacing (216 ± 2 beats/min), and 1:1 capture confirmed by electrocardiography. The dogs were then randomly assigned to one of four treatment protocols:

  1. Rapid-pacing-only: Dogs were given empty gelatin capsules during the pacing period of 28 days (n = 7).
  2. DMPI administration during rapid pacing: Dogs were administered DMPI, a dual-acting metalloprotease inhibitor (BMS-186716; Bristol-Myers Squibb, Princeton, NJ, U.S.A.) at a dose of 10 mg/kg, p.o., b.i.d., during the pacing period (n = 6).
  3. DMPI control: Dogs were administered DMPI at the same dose for 4 weeks without placement of pacing instrumentation (n = 6).
  4. Sham control: These dogs were instrumented and cared for in identical fashion to the groups described, with the exception of activation of the pacemaker and drug treatment (n = 6).

In all study groups, electrocardiograms and ventricular-pressure recordings were performed weekly during the 28-day pacing protocol to ensure proper operation of the pacemaker and the presence of 1:1 conduction. At the conclusion of the 28-day pacing protocol, the dogs were returned to the laboratory for terminal study, as described in the following section.

Dosage rationale

In preliminary studies, the dose of DMPI used in the long-term experiments was demonstrated to inhibit the angiotensin I (Ang I) pressor response (ACE inhibition). Specifically, dogs (n = 3) were administered the DMPI compound (10 mg/kg, p.o., b.i.d.) for 72 h to achieve steady-state plasma levels. After the morning dose on the fourth day, Ang I (100 ng/kg) was infused, and serial blood pressure measurements were taken. Compared with pre-DMPI values, the pressor response to intravenous infusion of Ang I was reduced by 87% at 1 h and 37% at 12 h after the morning DMPI dose. This specific DMPI compound was observed to potentiate ANP levels in primates, which is consistent with NEP inhibition (unpublished data, 1997). Finally, the dose of DMPI was selected minimally to affect systemic blood pressure in the normal state and therefore minimize the confounding influences of LV-loading conditions.

LV function measurements

Indices of LV systolic and diastolic function were obtained at baseline by using simultaneously recorded pressure and echocardiographic measurements previously described (18-20). All measurements were performed in a darkened room with the animal resting quietly in a sling. The arterial-access port was punctured with a 22-gauge Huber-point needle (Access Technologies, Skokie, IL, U.S.A.) connected to a fluid-filled catheter. Pressures from the fluid-filled aortic catheter were obtained by using an externally calibrated transducer (Statham P23ID; Gould, Oxnard, CA, U.S.A.). The ECG and pressure waveforms were recorded by using a multichannel recorder (TA4000; Gould, Irvine, CA, U.S.A.), as well as digitized on computer for subsequent analysis at a sampling frequency of 250 Hz (PO-NE-MAH, Storrs, CT, U.S.A.). Two-dimensional and M-mode echocardiographic studies (ATL Ultramark 7, 3.5 MHz transducer, Bothell, WA, U.S.A.) were used to image the LV from a right parasternal approach. LV volumes and ejection fraction were computed from the two-dimensional and M-mode echocardiographic recordings (18-20).

Neurohormonal measurements

To examine the relation between changes in neurohormonal status that accompany changes in LV function with long-term rapid pacing, blood samples were drawn at the conclusion of the LV-function study. With the animal resting quietly, 35 ml of blood was drawn from the arterial-access port into tubes containing EDTA (1.5 mg/ml), sodium azide (0.2 mg/ml), and aprotinin (1.15 TIU/ml). The blood samples were immediately centrifuged (2,000 g, 10 min, 4°C), the plasma decanted into separate tubes, frozen in a dry ice/methanol bath, and stored at −80°C until the time of assay. From these plasma samples, norepinephrine concentration, ANP levels, cGMP content, and plasma renin activity were determined. Plasma norepinephrine was measured by using high-performance liquid chromatography and normalized to pg/ml of plasma (21). For the ANP and cGMP assays, the plasma was first eluted over a cation exchange column (C-18 Sep-Pak; Waters Associates, Milford, MA, U.S.A.). Standardized radioimmunoassay procedures were performed to determine ANP concentrations, cGMP levels, and plasma renin activity (Peninsula Laboratories, Belmont, CA, U.S.A.). All plasma assays were performed in duplicate.

Myocyte isolation and myocardial sampling

After the completion of the protocols described earlier, the dogs were brought to the laboratory, and a final series of LV-function measurements and plasma samples was obtained. The animals were then anesthetized as described in the preceding section, a sternotomy performed, and the heart quickly extirpated and placed in a phosphate-buffered ice slush. The great vessels, atria, and right ventricle were carefully trimmed away, and the LV weighed. The region of the LV free wall incorporating the circumflex artery (5 × 5 cm) was excised and prepared for myocyte isolation. The posterior region of the LV free wall (4 × 4 cm) was snap frozen in liquid nitrogen for subsequent sarcolemmal preparation. The region of the LV free wall comprising the left anterior descending artery (3 × 5 cm) was cannulated and prepared for perfusion fixation.

Myocytes were isolated from the LV free wall by using methods described by this laboratory previously (22-25). In brief, the left circumflex coronary artery was perfused with a collagenase solution (0.5 mg/ml, Worthington, type II; 146 U/mg) for 35 min. The tissue was then minced into 2-mm sections and gently agitated. After 15 min, the supernatant was removed, filtered, and the cells allowed to settle. The myocyte pellet was then resuspended in Dulbecco's Modified Eagle's Medium: Nutrient Mixture F-12 (Gibco Laboratories, Grand Island, NY, U.S.A.). With this myocyte-isolation method, a high yield (75 ± 4%) of viable myocytes were routinely obtained for myocyte contractile function measurements, as described in the following section.

Myocardial structure analysis

The LV section for microscopic analysis was perfused with a buffered sodium cacodylate solution containing 2% paraformaldehyde, 2% glutaraldehyde solution (pH 7.4, 325 mOsm) for 20 min with a perfusion pressure of 100 mm Hg. Full-thickness LV samples (1 cm in thickness) were embedded in paraffin, sectioned at 5 μm in thickness, and stained with hematoxylin and eosin. These sections were imaged by using an epi-fluorescence illuminator with a rhodamine filter at a magnification of ×100. Myocytes in a cross-sectional orientation were digitized and analyzed by using an image-analysis system (Sigma Scan; Jandel Scientific, Corte Madera, CA, U.S.A.). Only those myocytes in which the nucleus was centrally located within the cell were digitized and analyzed to ensure that the short axis of the cardiocyte was perpendicular to the microscope objective. With this approach, myocyte cross-sectional area could be determined in situ.

Myocyte contractile function measurements

Isolated myocytes were placed in a thermostatically controlled chamber (37°C) fitted with a coverslip on the bottom for imaging on an inverted microscope (Sedival, Jena, Germany). The volume of the chamber was 2.5 ml, and it contained two stimulating platinum electrodes. The myocytes were imaged by using a ×20 long-working-distance objective. Myocyte contractions were elicited by field stimulating the tissue chamber at 1 Hz (S11; Grass Instruments, Quincy, MA, U.S.A.) by using current pulses of 5-ms duration and voltages 10% above contraction threshold. The polarity of the stimulating electrodes was alternated at every pulse to prevent the build-up of electrochemical by-products. Myocyte contractions were imaged by using a high-speed camera with noninterlaced scan rate of 240 Hz (GPCD60; Panasonic, Secaucus, NJ, U.S.A.). Myocyte motion signals were captured with the cell parallel to the video raster lines, and this video signal was input through an edge-detector system (Crescent Electronics, Sandy, UT, U.S.A.). The changes in light intensity at the myocyte edges were used to track myocyte motion (25). The distances between the left and right myocyte edges were converted into a voltage signal, digitized, and input into a computer (80386; Zenith Data Systems, St. Joseph, MI, U.S.A.) for subsequent analysis. Stimulated myocytes were allowed a 5-min stabilization period, after which contraction data for each myocyte were recorded from a minimum of 20 consecutive contractions. Parameters computed from the digitized contraction profiles included percentage shortening (%), peak shortening velocity (μm/s), peak relengthening velocity (μm/s), and total contraction duration (ms). After collection of baseline indices of myocyte function, measurements were then performed in the presence of 25 nM (−)isoproterenol (24).

Myocardial biochemical analysis

In light of the fact that abnormalities in β-adrenergic-receptor density and transduction have been reported with the development of pacing-induced LV dysfunction (17,23,24,26-30), our study examined β-receptor density and adenosine 3′,5′-cyclic monophosphate (cAMP) production with long-term rapid pacing in the presence and absence of DMPI administration. β-Adrenergic binding and function experiments were performed by using well-described methods (27-30). In brief, myocyte membranes were prepared by using ultracentrifugation techniques, and β-adrenergic-receptor antagonist binding studies were performed by using 10 concentrations of 25 μl [125I]cyanopindolol (ICYP; 74 Bq/mmol; Amersham Corp., Arlington Heights, IL, U.S.A.) from 0.02 to 1 nM. A standard Scatchard linear-regression analysis was performed on the amount of bound/free ligand with an r2 of >0.90 as the criterion for acceptability of the data. With this analysis, the maximal number of binding sites (Bmax), expressed as fmol/mg of protein, and the equilibrium dissociation constant Kd (nM) were computed (29,30). As an index of β-adrenergic-receptor system function, adenylate cyclase activity was determined by timed cAMP production in aliquots of 30-50 μg/100 μl of membrane preparation by using well-described methods (27,29). In addition to determining basal adenylate cyclase activity, cAMP production was measured in the presence of 10−3M (−)isoproterenol and 100 μM forskolin. These concentrations of isoproterenol and forskolin were shown previously to cause maximal adenylate cyclase activity in sarcolemmal preparations (29). Reactions were terminated by placing the tubes in boiling water and followed by centrifugation at 6,500 g for 5 min. The cAMP content of the supernatant was determined by using a competitive radiolabeled assay (cAMP 125I RIA; Advanced Magnetics Inc., Cambridge, MA, U.S.A.). Results were expressed as pmol cyclic AMP produced/mg sarcolemmal protein/min. All measurements were performed in duplicate.

Data analysis

Indices of LV and myocyte function were compared among the treatment groups by using analysis of variance. Analysis of the morphologic data was performed by using the average measurements obtained for each animal, and the groups were compared by using analysis of variance. If the analysis of variance revealed significant differences, pairwise tests of individual group means were compared by using Bonferroni probabilities (31). For comparisons of neurohormonal profiles between groups, the Wilcoxon and Mann-Whitney Rank-sum test was used (31). All statistical procedures were performed with the BMDP statistical software package (BMDP Statistical Software Inc., Los Angeles, CA, U.S.A.). Results are presented as mean ± standard error of the mean (SEM). Values of p < 0.05 were considered to be statistically significant.

RESULTS

LV function with long-term rapid pacing: effects of DMPI

All 25 dogs entered into the study were successfully studied. Terminal measurements of LV function in the treatment groups are summarized in Table 1. In the dual-acting metalloprotease inhibitor (DMPI) control group, LV function was not significantly different from that in sham controls. Although LV peak wall stress was increased in the drug control group, it did not reach statistical significance (p = 0.09). After 28 days of rapid pacing, ambient resting heart rate was significantly increased, LV end-diastolic volume and peak wall stress were increased by >70% compared with control values, and LV end-diastolic pressure was nearly 3 times control values. With long-term rapid pacing, LV ejection fraction was reduced to ∼50% of control values. With DMPI concomitantly administered during long-term rapid pacing, ambient resting heart rate was increased from control values but reduced from rapid-pacing-only values. DMPI concomitantly administered during long-term rapid pacing decreased mean arterial pressure from control values. Compared with rapid-pacing-only values, LV peak wall stress and ejection fraction were unchanged with DMPI concomitantly administered during long-term rapid pacing. LV end-diastolic pressure and LV end-diastolic volume were decreased from rapid-pacing-only values with DMPI concomitantly administered during rapid pacing, but remained increased when compared with controls. In terms of a relative change in LV end-diastolic volume from the baseline state, there was no significant change in the sham control or DMPI control groups. However, with long-term rapid pacing, LV end-diastolic volume increased by 177 ± 10% from baseline (p < 0.05). With DMPI concomitantly administered during rapid pacing, LV end-diastolic volume increased by 129 ± 6% from baseline (p < 0.05), but the increase was significantly lower than rapid-pacing-only values (p < 0.05). In the long-term rapid-pacing groups, LV mass normalized to body weight was unchanged from control values.

TABLE 1
TABLE 1:
LV function with long-term rapid pacing: effects of dual metalloprotease inhibition

Neurohormone levels: effects of DMPI

Data for plasma neurohormone levels are presented in Table 2. Plasma levels of norepinephrine in the DMPI control group were not significantly different from sham controls. With long-term rapid pacing, plasma norepinephrine was increased when compared with controls. DMPI concomitantly administered during long-term rapid pacing decreased plasma norepinephrine when compared with rapid pacing only, but this did not reach statistical significance (p = 0.4). Plasma ANP and plasma cGMP were not significantly different in the DMPI control group when compared with sham controls. Long-term rapid pacing increased both ANP and cGMP levels when compared with controls. However, DMPI concomitantly administered during long-term rapid pacing reduced both ANP and cGMP levels when compared with rapid-pacing-only. Long-term rapid pacing did not change plasma renin activity when compared with controls. Long-term DMPI administration increased plasma renin activity when compared with controls, and similarly, DMPI concomitantly administered during long-term rapid pacing increased plasma renin activity when compared with rapid-pacing-only.

TABLE 2
TABLE 2:
Plasma neurohormone levels with long-term rapid pacing: effects of dual metalloprotease inhibition

Myocyte geometry and contractility with long-term rapid pacing: Effects of DMPI

Frequency distribution of myocyte cross-sectional areas from LV myocardial sections are presented in Fig. 1. Mean myocyte cross-sectional area was 325 ± 5 μm2 in the sham control group and was decreased with long-term DMPI administration (276 ± 5 μm2; p < 0.05). Long-term rapid pacing also reduced mean myocyte cross-sectional area from control values (278 ± 5 μm2; p < 0.05). DMPI concomitantly administered during long-term rapid pacing further reduced mean myocyte cross-sectional area when compared with rapid-pacing-only values (251 ± 5 μm2; p < 0.05). LV myocytes were successfully isolated from all dogs. LV myocyte resting length was 152 ± 1 μm in the sham control group and was unchanged with long-term DMPI administration (148 ± 5 μm; p > 0.4). With long-term rapid pacing, LV myocyte resting length was increased from control values (178 ± 2 μm; p < 0.05). However, LV myocyte resting length was reduced with DMPI concomitantly administered during long-term rapid pacing from rapid-pacing-only values (159 ± 4 μm; p < 0.05).

FIG. 1
FIG. 1:
Frequency distribution of myocyte cross-sectional area from left ventricular (LV) myocardial sections perfusion fixed in situ. Myocyte cross-sectional area values were fitted to a gaussian distribution and are indicated by the solid lines. Dual metalloprotease inhibitor (DMPI) administration (BMS-186716; 10 mg/kg, p.o., b.i.d. for 28 days) as well as long-term rapid pacing (216 ± 2 beats/min for 28 days) reduced myocyte cross-sectional area compared with controls (p < 0.05). A further decline in LV myocyte cross-sectional area was observed with concomitant DMPI administration during rapid pacing when compared with rapid pacing only (p < 0.05). Overall mean values for cross-sectional area are presented in the Results section.

Steady-state basal contractile function for isolated LV myocytes was examined in >700 myocytes from each group and is summarized in Table 3. Baseline function in the DMPI control group was unchanged from sham controls. In the rapid-pacing-only group, the percentage shortening, shortening velocity, and relengthening velocity were significantly decreased compared with control values. With DMPI concomitantly administered during long-term rapid pacing, baseline contractile function improved when compared with rapid-pacing-only values. Thus DMPI concomitantly administered during long-term rapid pacing improved steady-state basal LV myocyte contractile performance.

TABLE 3
TABLE 3:
Isolated myocyte function with long-term rapid pacing: effects of dual metalloprotease inhibition

LV myocyte contractile function with β-adrenergic stimulation with 25 nM isoproterenol is summarized in Table 3. In all treatment groups, LV myocyte contractile function was increased from baseline with β-adrenergic stimulation. Myocyte contractile function after β-adrenergic receptor stimulation was unchanged in the DMPI control group compared with sham controls. However, compared with controls, myocyte contractile function in the presence of isoproterenol was reduced in the rapid-pacing-only group. With DMPI concomitantly administered during long-term rapid pacing, myocyte contractile function after β-receptor stimulation was similar to control values. In light of the fact that baseline myocyte contractile function was different in the treatment groups, the capacity of the myocyte to respond to an inotropic stimulus can be difficult to interpret. Therefore the absolute change in myocyte shortening velocity from baseline in the presence of isoproterenol was calculated for each cell. Results from this analysis are shown in Fig. 2. There was no difference in the absolute change in myocyte shortening velocity after β-receptor stimulation between the sham control and DMPI control groups. The absolute change in myocyte shortening velocity in the presence of isoproterenol in the rapid-pacing-only group was decreased compared with the control values. However, the absolute change in myocyte shortening velocity in the presence of isoproterenol was unchanged from control values with DMPI concomitantly administered during long-term rapid pacing. Therefore, DMPI concomitantly administered during long-term rapid pacing preserved the β-adrenergic response in isolated LV myocytes.

FIG. 2
FIG. 2:
Absolute change in myocyte-shortening velocity from baseline in the presence of 25 nM isoproterenol. Myocyte β-adrenergic response was significantly reduced in the rapid-pacing group (216 ± 2 beats/min for 28 days). Concomitant dual metalloprotease inhibitor (DMPI) administration (BMS-186716; 10 mg/kg, p.o., b.i.d.) during rapid pacing improved myocyte β-adrenergic response. *p < 0.05 vs. control vehicle; p < 0.05 vs. rapid pacing.

b-Receptor system: Effects of DMPI

The results of the analysis of β-receptor density and affinity and cAMP production in LV sarcolemmal preparations are presented in Table 4. In the DMPI control group, β-receptor density remained unchanged when compared with sham controls, but receptor affinity was increased. β-Receptor density decreased in both the rapid-pacing-only group and with DMPI concomitantly administered during long-term rapid pacing when compared with controls. cAMP production was markedly increased after β-receptor stimulation with isoproterenol and after direct adenylate cyclase activation with forskolin in control sarcolemmal preparations, and this response was increased in the DMPI control group when compared with sham controls. Long-term rapid pacing decreased cAMP accumulation when compared with control values. However, basal cAMP accumulation was unchanged from control values with DMPI concomitantly administered during long-term rapid pacing. Furthermore, cAMP accumulation in response to β-receptor stimulation was greater than control values with DMPI concomitantly administered during long-term rapid pacing. Therefore, DMPI concomitantly administered during long-term rapid pacing potentiated cAMP production in LV sarcolemmal preparations.

TABLE 4
TABLE 4:
Changes in β-adrenergic receptor and transduction with long-term rapid pacing: effects of dual metalloprotease inhibition

DISCUSSION

ACE inhibition was demonstrated to provide beneficial effects in chronic LV-pump dysfunction (13-15,17,32-35). Similarly, administration of an NEP inhibitor was shown to improve systemic hemodynamics in LV dysfunction (7-12,36-39). However, the effects of long-term inhibition of both ACE and NEP on LV and myocyte geometry and function with the development of LV failure remained unclear. Accordingly, Our study examined the effects of a dual-acting metalloprotease inhibitor (DMPI) on LV and myocyte geometry and function in a model of pacing-induced LV failure. The significant findings of this study were twofold. First, DMPI concomitantly administered during long-term rapid pacing reduced the degree of LV dilation and improved myocyte contractile function. However, this improvement in LV geometry and myocyte function was not translated into improved LV-pump performance. Second, DMPI concomitantly administered during long-term rapid pacing influenced β-adrenergic transduction through increased cAMP accumulation, which was translated into improved inotropic response at the level of the myocyte. To our knowledge, this is the first study to examine the long-term effects of DMPI administration on LV and myocyte geometry and function with the development of LV failure. However, a number of studies showed that long-term ACE inhibition improved LV ejection fraction after either coronary artery microembolization (33), aortic banding (34), or long-term rapid pacing (17). Furthermore, short-term inhibition of both ACE and NEP was shown to increase cardiac output in a model of pacing-induced LV failure (10). In a Syrian hamster model of cardiomyopathy, short-term administration of combined ACE and NEP inhibition improved LV-pump function (9). However, in both of these past short-term studies, systemic changes in mean arterial pressure and renal perfusion occurred. Thus the increased LV-pump function after a short-term infusion of combined ACE and NEP inhibitors, which was observed in these past studies, was likely due to favorable changes in LV loading conditions. Long-term rapid pacing causes significant LV dilation (17,18,22-24,26,28,29,40-43), and it was shown previously that concomitant ACE inhibition can decrease the LV dilation seen in this model (17). In our study, the degree of LV dilation was significantly reduced with DMPI concomitantly administered during long-term rapid pacing. A likely contributory factor for the reduction in LV dilation with DMPI treatment is through the ACE inhibitory component. In our study, DMPI concomitantly administered during long-term rapid pacing did not improve ejection fraction compared with rapid pacing only. However, it must be recognized that LV ejection fraction is influenced by a number of factors including preload, afterload, and contractile state. In light of the improved myocyte contractile function that occurred with DMPI treatment and rapid pacing, potential contributory mechanisms for the persistent reduction in LV ejection fraction include changes in LV loading conditions and geometry.

Neurohormonal levels

NEP is the enzyme primarily responsible for the degradation of ANP and is located in the lung and proximal tubules of the kidney (5,6). Whereas past studies showed that short-term NEP inhibition potentiates plasma ANP levels (7-12,36-39), the long-term effects of NEP inhibition on plasma ANP levels are not fully understood (8,12,36-39). For example, in a recent clinical trial, short-term NEP inhibition initially increased ANP levels, but this effect decreased over time with long-term inhibition (36). Willenbrock et al. (37) found that short-term NEP inhibition increased plasma ANP levels in rats, but ANP levels normalized over 30 days of long-term inhibition. Several past reports covered the effects of combined ACE and NEP inhibition on ANP levels (9,10,12). Trippodo et al. (9) demonstrated that combined ACE and NEP inhibition failed significantly to increase plasma ANP in cardiomyopathic hamsters. Seymour et al. (10) did not report an increase in plasma ANP with combined ACE and NEP inhibition after 3 weeks of long-term rapid pacing in dogs. In our study, long-term DMPI administration did not increase plasma ANP levels in DMPI controls. Furthermore, plasma ANP was reduced with DMPI concomitantly administered during long-term rapid pacing when compared with rapid-pacing-only. Consistent with the effect of long-term DMPI administration on plasma ANP levels, plasma levels of cGMP, the second messenger of ANP-receptor activation, failed significantly to increase in DMPI controls and was reduced with DMPI concomitantly administered during long-term rapid pacing when compared with rapid-pacing-only. Several potential contributory mechanisms may explain the reduction of plasma ANP and cGMP observed with long-term DMPI administration. First, long-term NEP inhibition may result in a new equilibrium between ANP production and degradation, which would lead to a compensatory shift in ANP and cGMP toward normal levels over time. A second contributory mechanism for the reduction in ANP observed in our study may be the effects of long-term ACE inhibition, a component of the dual-acting inhibitor, DMPI. Atrial stretch, which is mediated by increased atrial filling pressures, induces ANP synthesis and release (1,2,9,17,44). In our study, LV end-diastolic pressure was decreased with DMPI concomitantly administered during long-term rapid pacing when compared with rapid-pacing-only values and therefore may have reduced the stimulus for ANP release. However, characterization of the NEP inhibitory activity with short- and long-term administration of DMPI and mechanisms by which long-term DMPI treatment influences ANP synthesis and degradation warrant further investigation.

LV structure

To examine the cellular basis for changes in LV structure with long-term DMPI administration, this study examined the specific effects of long-term DMPI administration on myocyte cross-sectional area and resting length during the development of pacing-induced LV failure. Consistent with past studies, long-term rapid pacing reduced myocyte cross-sectional area and increased myocyte resting length (17,22,24,25,43). DMPI concomitantly administered during long-term rapid pacing reduced myocyte cross-sectional area and resting length from rapid-pacing-only values. Myocyte cross-sectional area also was reduced in the DMPI control group when compared with sham controls, thus suggesting that long-term DMPI administration influences myocardial growth. Myocyte cross-sectional area and resting length were shown to be modulated in a fashion similar to that observed in our study by concomitant ACE inhibition during long-term rapid pacing (17). Thus a likely contributory mechanism for the reduction in myocyte cross-sectional area and resting length is through the ACE-inhibitory component of this dual-acting compound. Past reports from this laboratory demonstrated that changes in the extracellular framework of the myocardium occurred with the development of pacing-induced LV failure (17,42,45). Specifically, collagen content and myocyte support are decreased with long-term rapid pacing (17,42). Additionally, a recent report by Kajstura et al. (43) suggested that side-to-side slippage and realignment of myocytes within the LV free wall may contribute to LV myocardial remodeling and subsequent dilation, which occurs with pacing-induced LV failure. In our study, DMPI concomitantly administered during long-term rapid pacing reduced the degree of LV dilation, myocyte cross-sectional area, and resting length. Taken together, these findings suggest that DMPI concomitantly administered during developing LV failure may have significant effects on myocyte geometry and alignment and extracellular support within the LV free wall, which in turn may influence global LV-pump performance.

Myocyte function

Consistent with past reports, long-term rapid pacing caused isolated myocyte contractile dysfunction (17,23,24,26,29). Previous reports from this laboratory demonstrated an improvement in isolated myocyte contractile function with concomitant ACE inhibition during long-term rapid pacing (17). In our study, DMPI concomitantly administered during long-term rapid pacing improved basal isolated myocyte contractile function to a greater degree than that seen with ACE inhibition alone (17). Although it remains speculative, a potential mechanism for this improvement may be an enhanced ACE inhibitory effect or inhibition of NEP or both. In a past report, long-term ACE inhibition was associated with improved LV-pump performance in addition to increased myocyte contractile function (17). In contrast to these findings, the increased myocyte contractile function with DMPI concomitantly administered during long-term rapid pacing was not translated into improved global LV-pump performance. A potential mechanism for this finding is that DMPI concomitantly administered during long-term rapid pacing leads to significant LV myocardial remodeling, which may, in turn, prevent the translation of myocyte contractile function into improved global LV-pump performance. Specific effects of LV myocardial remodeling on global LV-pump performance with DMPI concomitantly administered during long-term rapid pacing warrant future studies.

β-Adrenergic system

The development of pacing-induced LV failure is associated with a decreased β-adrenergic response in isolated myocytes (17,23,24,26-30). Mechanisms for reduced β-adrenergic response include neurohormonal activation and increased levels of circulating catecholamines with subsequent downregulation of β-adrenergic receptors (17,29,33). In our study the β-adrenergic response in isolated myocytes was preserved with DMPI concomitantly administered during long-term rapid pacing. Furthermore, long-term DMPI administration increased β-receptor and adenylate cyclase-mediated accumulation of cAMP within LV myocardial preparations. Because β-receptor density was not affected by long-term DMPI administration, changes in cAMP accumulation observed in our study are likely due to alterations in the intracellular signaling pathway. It was demonstrated that cGMP levels influence phosphodiesterase activity modulating cAMP degradation (11,44,46). Thus alterations in the cGMP intracellular signaling pathway will influence β-receptor-mediated cAMP levels. DMPI treatment in our study did not potentiate plasma cGMP levels, but the possibility exists that DMPI treatments altered intracellular cGMP levels. Thus long-term DMPI treatment may have reduced phosphodiesterase activity within LV myocytes, potentiating cAMP levels, leading to the preservation of β-adrenergic response in isolated myocytes. However, this issue remains speculative, and the effects of long-term DMPI administration on cAMP degradation warrant future investigations.

Summary

In a model of rapid pacing-induced LV failure, concomitant DMPI treatment significantly reduced the degree of LV dilation with no apparent effect on LV-pump function. At the level of the LV myocyte, long-term DMPI treatment with rapid pacing improved myocyte performance and β-adrenergic response. Thus the improvement in isolated myocyte contractile function was not translated into improved global LV-pump performance. The mechanisms by which improved myocyte contractility was not translated into a beneficial effect on LV pump function with DMPI treatment during rapid pacing remain speculative, but likely include significant changes in LV remodeling and loading conditions.

Acknowledgment: This study was supported by National Institutes of Health grant R29-HL-45024 (F.G.S.), a Basic Research Grant from Bristol-Myers Squibb Pharmaceutical Research Institute (F.G.S.), Thoracic Surgery Foundation for Research and Education (J.D.W.), and MUSC Post-Doctoral Research Award (B.R.H., J.D.W.). J.D.W. is a Nina S. Braunwald Research Fellow. F.G.S. is an Established Investigator of the American Heart Association. C.V.T. is an American Heart Association Research Fellow.

REFERENCES

1. Brandt RR, Wright SR, Redfield MM, Burnett JC. Atrial natriuretic peptide in heart failure. J Am Col Cardiol 1993;22:86A-92A.
2. Burnett JC, Kao PC, Hu DC, et al. Atrial natriuretic peptide elevation in congestive heart failure in the human. Science 1986;231:1145-7.
3. Winaver J, Hoffman A, Abassi Z, Haramati A. Does the heart's hormone, ANP, help in congestive heart failure? News Physiol Sci 1995;10:247-53.
4. Francis GS, Benedict C, Johnstone DE, et al. Comparison of neuroendocrine activation in patients with left ventricular dysfunction with and without congestive heart failure: a substudy of the Studies of Left Ventricular Dysfunction (SOLVD). Circulation 1990;82:1724-9.
5. Erdos EG, Skidgel RA. Neutral endopeptidase 24.11 (enkephalinase) and related regulators of peptide hormones. FASEB J 1989;3:145-51.
6. Kenny AJ, Stephenson SL. Role of endopeptidase 24.11 in the inactivation of atrial natriuretic peptide. FEBS Lett 1988;232:1-8.
7. Seymour AA, Asaad MM, Lanoce VM, Fennell SA, Cheung HS, Rogers WL. Inhibition of neutral endopeptidase 3.4.24.11 in conscious dogs with pacing induced heart failure. Cardiovasc Res 1993;27:1015-23.
8. Elsner D, Muntze A, Kromer EP, Riegger GA. Effectiveness of endopeptidase inhibition (candoxatril) in congestive heart failure. Am J Cardiol 1992;70:494-8.
9. Trippodo NC, Fox M, Natarajan V, Panchal BC, Dorso CR, Asaad MM. Combined inhibition of neutral endopeptidase and angiotensin converting enzyme in cardiomyopathic hamsters with compensated heart failure. J Pharmacol Exp Ther 1993;267:108-16.
10. Seymour AA, Asaad MM, Lanoce VM, Langenbacher KM, Fennell SA, Rogers WL. Systemic hemodynamics, renal function and hormonal levels during inhibition of neutral endopeptidase 3.4.24.11 and angiotensin-converting enzyme in conscious dogs with pacing-induced heart failure. J Pharmacol Exp Ther 1993;266:872-83.
11. Margulies KB, Barclay PL, Burnett JC. The role of neutral endopeptidase in dogs with evolving congestive heart failure. Circulation 1995;91:2036-42.
12. Bralet J, Marie C, Mossiat C, Lecomte JM, Gros C, Schwartz JC. Effects of alatriopril, a mixed inhibitor of atriopeptidase and angiotensin I-converting enzyme, on cardiac hypertrophy and hormonal responses in rats with myocardial infarction: comparison with captopril. J Pharmacol Exp Ther 1994;270:8-14.
13. The CONSENSUS trial study group. Effects of enalapril on mortality in severe congestive heart failure: results of the Cooperative North Scandinavian Enalapril Survival Study (CONSENSUS). N Engl J Med 1987;316:1429-35.
14. The SOLVD investigators. Effect of enalapril on survival in patients with reduced left ventricular ejection fractions and congestive heart failure. N Engl J Med 1991;325:293-302.
15. Pitt B. Use of converting enzyme inhibitors in patients with asymptomatic left ventricular dysfunction. J Am Coll Cardiol 1993;22:158A-61.
16. Smith JB, Lincoln TM. Angiotensin decreases cyclic GMP accumulation produced by atrial natriuretic factor. Am J Physiol 1987;253:C147-50.
17. Spinale FG, Holzgrefe HH, Mukherjee R, et al. Angiotensin-converting enzyme inhibition and the progression of congestive cardiomyopathy: effects on left ventricular and myocyte structure and function. Circulation 1995;92:562-78.
18. Tomita M, Spinale FG, Crawford FA, Zile MR. Changes in left ventricular volume, mass, and function during development and regression of supraventricular tachycardia induced cardiomyopathy: disparity between recovery of systolic vs diastolic function. Circulation 1991;83:635-44.
19. Laurenceau JL, Malergue MC. The essentials in echocardiography, M-mode and 2-dimensional imaging. Norwell, MA: Martinus Nijhoff 1981:64-70.
20. Zile MR, Tanaka R, Lindroth JR, Spinale FG, Carabello BA, Mirsky I. Left ventricular volume determined echocardiographically by using a constant LV epicardial long axis to short axis dimension ratio throughout the cardiac cycle. J Am Coll Cardiol 1992;20:986-93.
21. Goldstein DS, Feuerstein G, Izzo HL, Kopin IJ, Keiser H. Validity and reliability of liquid chromatography with electrochemical detection for measuring plasma levels of norepinephrine and epinephrine in man. Life Sci 1986;83:265-9.
22. Spinale FG, Zellner JL, Tomita M, Crawford FA, Zile MR. Relation between ventricular and myocyte remodeling with the development and regression of supraventricular tachycardia-induced cardiomyopathy. Circ Res 1991;69:1058-67.
23. Spinale FG, Clayton C, Tanaka R, et al. Myocardial Na+,K+-AT-Pase in tachycardia induced cardiomyopathy. J Mol Cell Cardiol 1992;24:277-94.
24. Spinale FG, Tempel GE, Mukherjee R, et al. Cellular and molecular alterations in the β-adrenergic system with cardiomyopathy induced by tachycardia. Cardiovasc Res 1994;28:1243-50.
25. Mukherjee F, Hewett K, Crawford FA, Spinale FG. Cell and sarcomere contractile performance from the same cardiocyte using video microscopy. J Appl Physiol 1993;74:2023-33.
26. Spinale FG, Holzgrefe HH, Mukherjee R, et al. Left ventricular and myocyte structure and function with recovery from tachycardia induced cardiomyopathy. Am J Physiol 1995;268:H836-47.
27. Kiuchi K, Shannon RP, Komamura K, et al. Myocardial β-adrenergic receptor function during the development of pacing induced heart failure. J Clin Invest 1993;91:907-14.
28. Roth DA, Kazushi U, Helmer GA, Hammond HK. Downregulation of cardiac guanosine 5′-triphosphate binding proteins in right and left ventricle in pacing induced congestive heart failure. J Clin Invest 1993;91:939-49.
29. Tanaka R, Fulbright BM, Mukherjee R, Burchell SA, Zile MR, Spinale FG. The cellular basis for the blunted response to β-adrenergic stimulation in supraventricular tachycardia induced cardiomyopathy. J Mol Cell Cardiol 1993;25:1215-33.
30. Bristow MR, Ginsburg R, Umans V, et al. β1- and β2-adrenergic-receptor subpopulations in nonfailing and failing human ventricular myocardium: coupling of both receptor subtypes to muscle contraction and selective β1-receptor down-regulation in heart failure. Circ Res 1986;59:297-309.
31. Steel RGD, Torrie JH. Principles and procedures of statistics: a biometrical approach. 2nd ed. New York: McGraw-Hill, 1980:1-623.
32. Poole-Wilson PA. Relation of pathophysiological mechanisms to outcome in heart failure. J Am Coll Cardiol 1993;22:22A-9.
33. Sabbah HN, Shimoyama H, Kono T, et al. Effects of long-term monotherapy with enalapril, metoprolol, and digoxin on the progression of left ventricular dysfunction and dilation in dogs with reduced ejection fraction. Circulation 1994;89:2852-9.
34. McDonald KM, Garr M, Carlyle PF, et al. Relative effects of α1-adrenoceptor blockade, converting enzyme inhibitor therapy, and angiotensin II subtype 1 receptor blockade on ventricular remodeling in the dog. Circulation 1994;90:3034-46.
35. Weinberg EO, Schoen FJ, George D, et al. Angiotensin converting enzyme inhibition prolongs survival and modifies the transition to heart failure in rats with pressure overload hypertrophy due to ascending aortic stenosis. Circulation 1994;90:1410-22.
36. Richards AM, Wittert GA, Crozier IG, et al. Chronic inhibition of endopeptidase 24.11 in essential hypertension: evidence for enhanced atrial natriuretic peptide and angiotensin II. J Hypertens 1993;11:407-16.
37. Willenbrock R, Scheuermann M, Höhnel K, Friedrick CL, Dietz R. Acute and chronic neutral endopeptidase inhibition in rats with aortocaval shunt. Hypertension 1996;27:1259-66.
38. Jin H, Mathews C, Chen YF, et al. Effects of acute and chronic blockade of neutral endopeptidase with SCH 34826 on NaCl-sensitive hypertension in spontaneously hypertensive rats. Am J Hypertens 1992;5:210-8.
39. Ogihara T, Rakugi H, Masuo K, Yu H, Nagano M, Mikami H. Antihypertensive effects of the neutral endopeptidase inhibitor SCH 42495 in essential hypertension. Am J Hypertens 1994;7:943-7.
40. Howard RJ, Stopps TP, Moe GW, Gotlieb A, Armstrong PW. Recovery from heart failure: structural and functional analysis in a canine model. Can J Physiol Pharmacol 1988;66:1505-12.
41. Damiano RJ, Tripp HF, Asano T, Small KW, Jones RH, Lowe JE. Left ventricular dysfunction and dilation resulting from chronic supraventricular tachycardia. J Thorac Cardiovasc Surg 1987;94:135-43.
42. Spinale FG, Tomita M, Zellner JL, Cook JC, Crawford FA, Zile MR. Collagen remodeling and changes in LV function during development and recovery from supraventricular tachycardia. Am J Physiol 1991;261:H308-18.
43. Kajstura J, Zhang X, Liu Y, et al. The cellular basis of pacing-induced dilated cardiomyopathy: myocyte cell loss and myocyte cellular reactive hypertrophy. Circulation 1995;92:2306-17.
44. Clemo HF, Baumgarten CM. cGMP and atrial natriuretic factor regulate cell volume of rabbit atrial myocytes. Circ Res 1995;77:741-9.
45. Spinale FG. Pacing tachycardia-induced congestive heart failure. Heart Failure 1995;11:219-32.
46. Lincoln TM. Cyclic CMP: biochemistry, physiology, and pathophysiology. Austin: RG Landes Company, 1994:7-27.
Keywords:

Cardiomyopathy; Neutral endopeptidase; Angiotensin-converting enzyme; Left ventricular dysfunction; Myocyte contractility

© 1998 Lippincott Williams & Wilkins, Inc.