Role of Angiotensin-Converting Enzyme and Neutral Endopeptidase in Flow-Dependent RemodelingKorshunov V.A.a · Massett M.P.a · Carey R.M.b · Berk B.C.a
aCenter for Cardiovascular Research and Department of Medicine, University of Rochester, Rochester, N.Y., and bDepartment of Internal Medicine, University of Virginia Health System, Charlottesville, Va., USA Corresponding Author
Omapatrilat inhibits neutral endopeptidase (NEP) and angiotensin-converting enzyme (ACE). We compared the effects of omapatrilat (40 mg/kg/day, p.o.) to fosinopril (40 mg/kg/day, p.o.) on flow-induced vascular remodeling in New Zealand genetically hypertensive (GH) rats. Both drugs equally reduced blood pressure (BP) initially, but systolic BP and pulse pressure were reduced more by omapatrilat after 1 week. Carotid remodeling was induced by partial ligation of the left common carotid artery (LCA). There was little remodeling in untreated GH rats – measured as outer diameter to body weight (OD/BW vs. before ligation): 97 ± 1% of initial LCA (low flow) and 107 ± 3% of initial right common carotid artery (RCA, high flow). In contrast, OD/BW increased to 118 ± 5% (p < 0.05) of initial RCA after omapatrilat versus 108 ± 2% (p = 0.96) after fosinopril. The major change was increased RCA lumen area which was significantly larger in omapatrilat-treated animals (127% vs. control) than fosinopril-treated animals (103% vs. control). The increase in outward remodeling after omapatrilat treatment correlated weakly with vascular cGMP levels and decreased systolic BP. The results suggest that dual inhibition of NEP/ACE may have greater effects than ACE inhibition alone on vessel remodeling in hypertension.
Copyright © 2004 S. Karger AG, Basel
A novel approach for the treatment of hypertension is the use of dual inhibitors of neutral endopeptidase (NEP) and angiotensin-converting enzyme (ACE), such as omapatrilat . Recently, it was reported that omapatrilat may exhibit advantages over ACE inhibitors in the treatment of patients with congestive heart failure . Based on the mechanism of action, the most beneficial effects of omapatrilat are expected to involve improved endothelial function and vascular structure. For example, chronic omapatrilat treatment improved endothelium-dependent vasodilation of resistance arteries in the stroke-prone spontaneously hypertensive rat . Furthermore, in Dahl salt-sensitive rats omapatrilat normalized endothelium-dependent relaxations and increased cGMP levels in resistance arteries compared with captopril .
The ability of vessels to remodel has been identified as a pathogenic factor for atherosclerosis, hypertension and restenosis. Outward remodeling (a radial enlargement of vessel diameter) may prevent ischemia in atherosclerosis by preserving lumen size and represents a strategy that should improve patient outcomes . Enlargement of the lumen diameter can occur in response to chronic increases in blood flow and shear stress . We have developed a highly reproducible model of flow-induced vascular remodeling in the rat carotid artery . Our previous data suggest that elevated blood flow in the carotid artery increases protein levels of endothelial nitric oxide synthase (eNOS) [7, 8, 9]. Thus, flow-induced outward remodeling may be due to augmented NO production and increased cGMP in the target vascular smooth muscle cells. Because NEP/ACE inhibitors can improve endothelial function [3, 4], these agents may also augment outward remodeling. Thus, the purpose of the present study was to investigate the effects of NEP/ACE inhibition by omapatrilat compared with ACE inhibition by fosinopril on flow-induced vascular remodeling.
Male New Zealand genetically hypertensive (GH) rats (7–9 weeks old; body weight, ∼123 g) were used in accordance with the guidelines of the National Institutes of Health for the Care and Use of Laboratory Animals, and the study was approved by the University of Rochester Animal Care Committee. The animals were kindly provided by Dr. Eugenie Harris (University of Otago, Dunedin, New Zealand).
Radiotelemetry implants (TA11PA-C20, Data Science Int.) were inserted as previously described elsewhere . Briefly, the animals were anesthetized with an intraperitoneal injection of a cocktail of ketamine (60 mg/kg) and xylazine (5 mg/kg) and maintained at 37°C on a heating pad. The nonoccluding catheter with an antithrombogenic tip was implanted into the abdominal aorta with the tip just below the renal arteries and fixed in place with tissue adhesive. The implants were immobilized by suturing to the ventral abdominal wall, and wounds were closed by suture under aseptic conditions. The animals were allowed to recover for 10 days prior to experiments and housed individually under specific pathogen-free conditions at 12-hour light/dark cycles (lights on at 6 a.m., lights off at 6 p.m.). The cages were placed on receivers, and pulsatile arterial blood pressure (BP) and heart rate (HR) were recorded for a 10-second interval every 5 min during the experiment. Mean arterial blood pressure (MAP) was derived from the waveform by Dataquest A.R.T. software (Data Science Int.), version 1.1. The original MAP and HR were averaged for every hour. Data obtained 2 days prior to treatment were averaged and used for baseline (fig. 1). Changes in parameters were calculated for every hour by subtraction from baseline 24-hour data.
Fig. 1. Baseline (before treatment) 24-hour circadian rhythms of systolic BP (a), diastolic BP (b), MAP (c), PP (d), HR (e) and activity (f) recorded by radiotelemetry in New Zealand GH rats. GH rats have impaired circadian regulation (12-hour light-dark cycle) of BP, but have normal HR and activity. There were no statistically significant differences in baseline parameters among experimental groups. Values are means ± SEM. n = Number of animals per group. Lights on at 6 a.m., lights off at 6 p.m.
The study group was randomly divided into three subgroups: the control (n = 7), fosinopril [an ACE inhibitor (n = 10)], or omapatrilat [a dual NEP/ACE inhibitor (n = 9)] groups. Control animals received 5% solution of sodium bicarbonate (10 ml/kg) by gavage every day for 1 week. Fosinopril and omapatrilat (Bristol-Myers Squibb) were given 40 mg/kg per day solubilized in 5% solution of sodium bicarbonate by gavage. Preliminary experiments were used to determine the dose of fosinopril (20 and 40 mg/kg) that lowered MAP to the same degree as omapatrilat (40 mg/kg per day).
On the first day of treatment, the animals were anesthetized. The outer diameter (OD) of the left (LCA) and right common carotid arteries (RCA) were measured at three sites over 1–2 cm length using a microscope equipped with a reticule, and the average of three measurements was recorded. The blood flow was also measured on both carotids using an ultrasonic transit-time flowmeter (Transonic Systems). The left external and internal carotid arterial branches were isolated and ligated with 6-0 silk suture. The OD and blood flow measurements were repeated. The neck incision was closed, and rats were allowed to recover. Treatment continued on the next day. Bromodeoxyuridine (BrDU) tablets were placed subcutaneously 24 h prior to sacrifice under anesthesia in aseptic conditions. The measurements of OD and blood flow of both common carotids were performed as described above on the last day of the treatment (termination). Tissue samples (blood and thoracic aorta) were harvested and immediately frozen in liquid nitrogen for cGMP analysis .
All animals were perfusion fixed for 5 min with 10% paraformaldehyde in sodium phosphate buffer (pH 7.0) as previously described . The LCA and RCA were harvested and embedded in paraffin. All cross-sections were stained with hematoxylin and eosin. Selected samples were evaluated using BrDU antibody with hematoxylin counterstain . Elastin was evaluated with van Gieson stain (elastic stain, Chromaview, Richard-Allan Scientific). Digitized images of cross-sections were analyzed using NIH Image 1.62 software. Five or more sections were analyzed, and the mean for each animal was calculated. The density of BrDU-positive cells was calculated as the number of positive cells per media area (counts × 10–6 μm2).
The mean shear stress (τ) was calculated as described elsewhere . During the course of the experiment, all rats grew as evidenced by an increase in body weight. Animals gained approximately 5 g. Therefore, to evaluate the changes in vascular diameter independent of growth, changes in OD were normalized for body weight at termination.
All results are reported as means ± SEM. All statistical tests were done with Statview for MacIntosh, version 5.0.1. Differences between groups were analyzed by means of a repeated-measures one-way ANOVA; cardiovascular changes over time between groups were analyzed by two-way ANOVA and followed by Fisher’s post-hoc test. The level of p < 0.05 was regarded as significant.
The body weight of New Zealand GH rats at termination was not different across groups (∼180 g). Complete baseline telemetry data are presented in figure 1. GH rats exhibited abnormal light/dark regulation of BP, but had a normal regulation of activity and HR (fig. 2). Anti-hypertensive treatment of GH rats resulted in significant reductions in BP with both omapatrilat and fosinopril (fig. 3). We chose concentrations that would yield similar acute reductions in BP (within 24 h). However, over the course of the experiment we observed that MAP decreased to a greater extent with omapatrilat than with fosinopril (fig. 3c). The differences between drugs in MAP were due primarily to a continuous decrease of absolute systolic BP for the last 3 days as shown for day 6 (fig. 4). However, the maximal decrease in systolic BP (calculated by subtraction from the same day level of systolic BP before treatment) was not significantly different between fosinopril (–15 ± 3 mm Hg) and omapatrilat (–22 ± 4 mm Hg) even on the 6th day. The pulse pressure (PP) also decreased significantly in the omapatrilat group compared with control and fosinopril groups (fig. 3d). There were no significant differences in HR and moving activity among experimental groups (fig. 3e, f). Thus, both drugs equally decreased BP initially, but systolic BP and PP were reduced during 1-week treatment with omapatrilat to a greater extent than fosinopril in GH rats.
Fig. 2. Average 12-hour baseline (before treatment) of MAP (a), HR (b) and activity (c) recorded by radiotelemetry in GH rats. Data are from all GH rats included in the study (n = 26). GH rats have impaired circadian organization (12-hour light-dark cycle) of BP, but normal regulation of HR and activity. Day, 6 a.m. to 6 p.m.; night, 6 p.m. to 6a.m. Values are means ± SEM; * p < 0.05 vs. control (Student’s t test).
Fig. 3. Effects of antihypertensive treatment on systolic BP (a), diastolic BP (b), MAP (c), PP (d), HR (e) and activity (f) recorded by radiotelemetry in New Zealand GH rats. One-week antihypertensive treatment of GH rats resulted in significant reduction in BP, which was greater in the omapatrilat group. There were no significant changes in HR and activity after treatment. Time scale is in 12-hour increments. Values are means ± SEM; * p < 0.05 vs. control (ANOVA); ** p < 0.05 compared with fosinopril (ANOVA). n = Number of animals per group.
Fig. 4. Absolute values of systolic (■) and diastolic (&karsi020;) BP recorded by radiotelemetry in GH rats at 6 a.m. on the 7th day of treatment. Omapatrilat significantly reduced systolic BP compared with control and fosinopril before the last gavage. Values are means ± SEM; * p < 0.05 vs. control (ANOVA); ** p < 0.05 vs. fosinopril (ANOVA).
In response to ligation, blood flow significantly decreased (–90%) in the LCA and increased (+70%) in the contralateral side 1 week after ligation (table 1). Shear stress also significantly decreased in the LCA, but remained unchanged in the RCA. There was a trend toward decreased shear stress in the RCA for the omapatrilat group, but this was not statistically significant (table 1).
Table 1. Mean blood flow and shear stress before (initial) and 1 week after ligation (termination) in GH rats
Despite large alterations in blood flow, there were small changes in vascular remodeling in untreated, control GH rats measured as ΔOD/BW: 97 ± 1% at low flow, and 107 ± 3% at high flow (fig. 5). Treatment did not significantly change vascular remodeling in the low flow vessel (LCA). However, omapatrilat-treated animals showed a significantly greater response to increased flow (RCA) than control or fosinopril-treated rats. The ΔOD/BW increased to 118 ± 5% (p < 0.05) in omapatrilat rats versus 108 ± 2% (p = 0.97) in the fosinopril rats and 107 ± 3% in control rats. Morphologic examination showed the RCA (increased flow) from the omapatrilat group was larger compared with either control or fosinopril (fig. 6). We performed correlation analyses between changes in OD/BW and BP variables (table 2). We found no significant correlation, although changes in systolic BP (R2 = 0.27) and diastolic BP (R2 = 0.22) were more highly correlated than PP (R2 = 0.02).
Fig. 5. OD changes in carotid arteries normalized by body weight after 1-week ligation in control and fosinopril- or omapatrilat-treated GH rats. Omapatrilat significantly increased outward remodeling (ΔOD/BW) compared with control and fosinopril only in the right carotid. Values are means ± SEM; * p < 0.05 vs. control (ANOVA); ** p < 0.05 vs. fosinopril (ANOVA). n = Number of animals per group.
Fig. 6. Verhoeff’s van Gieson staining of the RCA from control (a), fosinopril- (b) and omapatrilat- (c) treated GH rats 1 week after ligation. There were no areas of apoptosis or neointima formation in carotids after ligation. Elastic fibers are black in inserts. Light microscope magnification is×4. Magnification bar is 200 μm. Insert magnification is ×60.
Table 2. Correlations between outward vascular remodeling (ΔOD/BW) and changes in blood pressure variables after treatment of GH rats
Morphometric analyses of the vessel components (fig. 6) showed the major finding to be a significantly larger RCA lumen area in omapatrilat-treated animals (127% vs. control) compared to fosinopril-treated animals (103% vs. control). In addition, the ratio of RCA lumen area to that of the LCA lumen area was significantly larger only in the omapatrilat group (fig. 7). Media area also significantly increased in the RCA from animals treated with omapatrilat compared with control and fosinopril-treated rats (table 3). Based on the increase in lumen and media areas in the RCA, the vessel area of omapatrilat-treated animals showed a tendency to be larger (∼15%) than in fosinopril-treated or control rats. However, because the adventitia area was smallest in this group compared with control or fosinopril, the difference in vessel area was not significant (table 3). Histological examination showed that neither RCA (fig. 6) nor LCA (data not shown) had neointima formation or evidence of apoptosis in any experimental group. Taken together, omapatrilat increased outward remodeling in response to high flow largely by an increase in lumen area in GH rats.
Fig. 7. Ratio of RCA/LCA from morphometry data after treatment in GH rats. The lumen ratio area was significantly larger only in the omapatrilat group. Values are means ± SEM; * p < 0.05 compared with control (ANOVA); ** p < 0.05 compared with fosinopril (ANOVA). n = Number of animals per group.
Table 3. Morphometric analyses of vessel cross-sections from control, fosinopril and omapatrilat-treated GH rats
To investigate mechanisms by which omapatrilat enhanced outward remodeling we analyzed cGMP levels in the plasma and aorta, and cell proliferation in the carotids. Both fosinopril and omapatrilat increased plasma cGMP level to a similar extent (fig. 8a). Because of the small amount of tissue and relatively low cGMP levels we measured cGMP in the aorta rather than the carotid. The aortic cGMP exhibited a trend to be higher in omapatrilat-treated animals compared with control or fosinopril-treated rats after 1 week of treatment although this was not statistically significant (fig. 8b). However, previous experiments  found a significant increase in cGMP after omapatrilat therapy. Cell proliferation was reduced (∼4-fold) in both the RCA and LCA after fosinopril and omapatrilat treatment. The inhibition of proliferation did not differ significantly for the two drugs (fig. 9).
Fig. 8. The level of cGMP in plasma (a) and thoracic aorta (b) of treated GH rats. &karsi020; = Controls (n = 2); ▧ = fosinopril (n = 4); ■ = omapatrilat (n = 3). There were no significant changes in plasma or aortic cGMP after treatment.
Fig. 9. Density of BrDU-positive cells per media area in carotid arteries from control, fosinopril- and omapatrilat-treated GH rats. Proliferation was significantly decreased in treated animals. Values are means ± SEM; * p < 0.05 vs. control. n = Number of animals per group; n = 5 for the RCA in the fosinopril group because one cross-section was damaged during processing.
The major finding of this study is that omapatrilat augmented outward remodeling of the carotid artery in response to increased flow to a significantly greater extent than fosinopril. Despite a similar decrease in MAP after treatment, omapatrilat continuously reduced systolic BP and PP compared with fosinopril. Only omapatrilat-treated animals exhibited increases in lumen and media areas in the high-flow RCA. Also, the ratio of RCA lumen area to LCA lumen area was significantly larger in the omapatrilat group. Outward remodeling correlated better with the decrease in systolic BP than changes in PP. It appears that multiple mechanisms explain the differences between omapatrilat and fosinopril. It is unlikely that inhibition of proliferation contributed to the difference in remodeling, since proliferation was attenuated comparably by both agents. These results suggest that the anti-proliferative effects of omapatrilat and fosinopril were mostly mediated by ACE inhibition. The finding that steady-state levels of cGMP were similar in both plasma and aorta (fig. 8) after treatment with omapatrilat and fosinopril suggests that differences in cGMP may not explain the increased remodeling observed with omapatrilat. Instead we suggest that the differing effects of omapatrilat (a dual NEP/ACE inhibitor) compared to fosinopril (an ACE inhibitor) on outward remodeling are due primarily to the augmented ability of omapatrilat as an NEP inhibitor to increase peptides metabolized by NEP, and secondarily to increased cGMP levels compared with ACE inhibition alone by fosinopril .
The mechanism(s) underlying outward vascular remodeling are unknown. However, it was shown that increased flow and shear stress are associated with increases in lumen diameter (outward remodeling) . Previously, we reported that elevated blood flow in the RCA after unilateral ligation increased protein levels of eNOS [7, 8, 9]. This suggests that one possible mechanism of outward vascular remodeling is to increase the level of endogenous NO. GH rats are reportedly more sensitive than either its normotensive control strain or the spontaneously hypertensive rat strain to changes in the synthesis of NO . Therefore, we propose that the beneficial effects of omapatrilat on remodeling in GH rats compared with fosinopril may be due in part to increased production or bioavailability of NO. Alternatively, because NO generates cGMP, the effect of omapatrilat may be secondary to increased cGMP (although the increase in cGMP was small) (fig. 8).
Inhibition of NEP is a new approach for treating high BP and other cardiovascular diseases . However, the utility of selective NEP inhibitors in the treatment of cardiovascular diseases is currently in question, because selective NEP inhibitors induced neurohormonal activation in hypertensive patients [17, 18]. When administered by gavage (40 mg/kg/day), as in the present study, omapatrilat caused complete inhibition of renal  and cardiac  NEP and ACE in spontaneously hypertensive rats. Because arterial BP affects vascular remodeling (table 2), we used dosages of omapatrilat and fosinopril that caused equivalent changes initially (during the first 3 days) in MAP (fig. 3). In fact, the maximal decrease in systolic BP (calculated by subtraction from the same day level of systolic BP before treatment) was not significantly different between drugs even on the last day of treatment. However, omapatrilat exhibited a greater reduction of systolic BP and PP than fosinopril for the last 3 days of treatment (fig. 3). The BP-lowering effect of omapatrilat is attributed to its actions on two enzymes: NEP and ACE. Dual inhibition of NEP and ACE by omapatrilat not only leads to inhibition of the renin-angiotensin system, but also to accumulation of natriuretic factors and vasodilatory peptides (e.g. ANP, BNP, bradykinin, adrenomedullin) . Both natriuretic peptides and bradykinin can enhance eNOS activity and increase levels of NO . Natriuretic peptides can also increase cGMP independently of eNOS via direct effects on particulate guanylate cyclase [22, 23]. As a result, inhibition of both NEP and ACE by omapatrilat can significantly improve endothelium-dependent relaxation and increase cGMP levels . However, dual inhibition of NEP and ACE may be more effective than ACE inhibition alone in increasing levels of vasodilatory peptides . It is important to note that there may be positive remodeling effects of these peptides that are independent of changes in cGMP, as shown for the cardioprotective effect of omapatrilat on hypertensive dogs .
In contrast to their disparate effects on outward vascular remodeling, both omapatrilat and fosinopril significantly reduced proliferation in the media (fig. 9) in the left carotid (low flow) compared with controls. This reduction may be a consequence of inhibiting both the growth-promoting effects of angiotensin II and enhancing the anti-proliferative effects of cGMP [26, 27]. Omapatrilat and fosinopril also attenuated proliferation in the media of the RCA to varying degrees. Although this attenuation was statistically significant only in the omapatrilat-treated rats, vessels from fosinopril-treated animals showed a trend toward decreased proliferation. The disparity between changes in proliferation and media area may be related to temporal differences in DNA synthesis measured by BrDU and structural changes. Previous data in balloon-injured vessels indicate that BrDU staining peaks 2–3 days after injury, while structural changes occur more slowly . Matrix hypertrophy may also have contributed to the increase in media area and this would not be reflected by changes in the number of BrDU-stained nuclei . The decrease in proliferation in media area and increases in lumen and media areas in the RCA coupled with no significant change in vessel wall ultrastructure (fig. 6), media thickness (data not shown) or right/left media area ratio (fig. 7) of omapatrilat-treated rats indicate that eutrophic outward remodeling occurred in these rats.
In conclusion, this report provides the first evidence that dual inhibition of NEP and ACE by omapatrilat augments outward remodeling in response to high flow to a greater extent than fosinopril. A possible mechanism is that inactivation of NEP prevents degradation of vasodilator peptides, which may be important for endothelium and smooth muscle cell function. Our results suggest that omapatrilat may be more effective than fosinopril in promoting outward remodeling in essential hypertension, and consequently, improve clinical outcomes.
This study was supported by Bristol-Myers Squibb and NIH grant HL-62826 to B.C.B. M.P.M. is an AHA Scientist Development Grant awardee. The authors would like to thank Benjamin Lee for surgical assistance, and Mary Georger and David Nagel for help in performing morphological measurements.
Dr. Bradford C. Berk
University of Rochester, Center for Cardiovascular Research
601 Elmwood Avenue
Rochester, NY 14642 (USA)
Tel. +1 585 273 1946, Fax +1 585 273 1497, E-Mail Bradford_Berk@URMC.rochester.edu
Received: July 16, 2003
Accepted after revision: September 30, 2003
Published online: March 3, 2004
Number of Print Pages : 9
Number of Figures : 9, Number of Tables : 3, Number of References : 28
Journal of Vascular Research (Incorporating International Journal of Microcirculation)
Founded 1964 as Angiologica by M. Comèl and L. Laszt (1964–1973) continued as Blood Vessels by J.A. Bevan (1974–1991)
Official Journal of the European Society for Microcirculation
Vol. 41, No. 2, Year 2004 (Cover Date: March-April 2004)
Journal Editor: U. Pohl, Munich
ISSN: 1018–1172 (print), 1423–0135 (Online)
For additional information: http://www.karger.com/jvr