OUP user menu

ACE inhibition promotes upregulation of endothelial progenitor cells and neoangiogenesis in cardiac pressure overload

Patrick Müller, Andrey Kazakov, Philippe Jagoda, Alexander Semenov, Michael Böhm, Ulrich Laufs
DOI: http://dx.doi.org/10.1093/cvr/cvp123 106-114 First published online: 20 April 2009


Aims Inhibition of the angiotensin-converting enzyme (ACE) prevents maladaptive cardiac remodelling. Endothelial progenitor cells (EPC) from the bone marrow contribute to endothelial repair and neovascularization, effects that are potentially important during cardiac remodelling. We hypothesized that ACE inhibitors may exert beneficial effects during pressure-induced myocardial hypertrophy by regulating progenitor cell function.

Methods and results In C57/Bl6 mice, development of cardiac hypertrophy induced by transaortic constriction (TAC) for 5 weeks was reduced by ramipril, 5 mg/kg p.o., independent of blood pressure lowering. Ramipril prevented TAC-induced apoptosis of cardiac myocytes and endothelial cells. On day 1 after TAC, upregulation of Sca-1pos/KDRpos EPC was observed, which was further increased by ramipril. EPC were persistently elevated in the TAC mice receiving vehicle treatment but not in the ramipril group after 5 weeks. These effects were independent of hypoxia-inducible factor-1α mRNA and protein expression. The ACE inhibitor but not TAC improved the migratory capacity of DiLDLpos EPC. Increased cardiac afterload induced upregulation of extracardiac neoangiogenesis. This effect was enhanced by ACE inhibition. Ramipril but not TAC markedly increased cardiac capillary density determined by the ratio of CD31pos cells to cardiomyocytes. Bone marrow transplantation studies revealed that TAC increased the percentage of bone marrow-derived GFPpos endothelial cells in the myocardium, and ramipril made this effect more pronounced.

Conclusions ACE inhibition prevents pressure-induced maladaptive cardiac hypertrophy and increases intra- and extracardiac neoangiogenesis associated with the upregulation of EPC and amelioration of EPC migration. The regulation of progenitor cells from the bone marrow identifies a novel effect of ACE inhibitors during cardiac remodelling.

  • Angiotensin-converting enzyme-inhibition
  • Endothelial progenitor cells
  • Neoangiogenesis
  • Transaortic constriction

1. Introduction

Angiotensin-converting enzyme-inhibitors (ACE-I) block the conversion of angiotensin I to angiotensin II and inhibit the breakdown of bradykinin. The well-documented antihypertensive and antihypertrophic properties of ACE-I are associated with beneficial cardiovascular effects.14 Recent data suggested potential pro-angiogenic effects of ACE inhibitors.5 Whether and how ACE inhibitors could potentially contribute to myocardial capillarization is still a matter of debate.6,7

Endothelial progenitor cells (EPC) play an important role during angiogenesis and vascular repair.8 In patients with coronary artery disease, the quantity of circulating CD34posKDRpos EPC predicts cardiovascular events and death from cardiovascular causes.9 The number and function of EPC are regulated by interventions such as physical activity and lipid lowering with statins, which increase the number and migratory function of EPCs.10,11 Vascular risk factors such as hypertension, hyperlipidaemia, or diabetes may reduce their number and function.1214 Hypertrophy and remodelling of cardiac muscle occurring during pathological cardiac load result in contractile dysfunction and heart failure, which are associated with high morbidity and mortality. The mechanisms leading to myocardial hypertrophy and therefore the effects of medical therapies to ameliorate pathological remodelling are still incompletely understood. Research was focused primarily on the ameliorating function of cardiomyocytes. But during the development of myocardial hypertrophy, a mismatch between the ratio of capillaries and cardiomyocytes develops, indicating that the regulation of angiogenesis may also play an important role in cardiac hypertrophy.15

In the heart, increased afterload leads to the upregulation of EPCs from the bone marrow, which may contribute to myocardial angiogenesis,16 but the effects are not able to compensate for the capillary–myocyte mismatch developing during and contributing to the progression of myocardial hypertrophy.15,16 However, these data show that myocardial afterload induces systemic regulation of bone marrow-derived progenitor cells which are then potentially able to exert intra-myocardial vascular effects and therefore may represent a novel target for antihypertrophic treatment. We therefore hypothesized that ACE inhibitors—which are used to prevent myocardial hypertrophy in hypertension—may exert part of their protective effects in cardiac remodelling processes by enhancing cardiac angiogenesis through affecting EPC numbers and function.

2. Methods

2.1 Animals, treatment with ramipril, and transaortic constriction

Protocols are described in detail in Supplementary material online. Shortly, animals received, at 10 weeks of age, either ramipril at a dose of 5 mg per kg body weight17 in the form of pulverized tablets (Aventis, Germany) introduced into the diet or in the form of a matched placebo diet (sniff, Germany). Three days after start with ramipril treatment or placebo diet, transaortic constriction (TAC) was performed. Control mice underwent a sham operation. After 5 weeks, animals were sacrificed, and after LV pressure measurements, hearts were rapidly excised. Hearts were partly snap-frozen and stored at −80°C, partly fixed and embedded in paraffin. Blood and bone marrow were sampled, as well as the spleen, for further preparation.

2.2 Bone marrow transplantation

Six-week-old C57Bl/6-Tg(ACTbEGFP)1Osb mice (Jackson Laboratory) were killed and marrow was obtained from the long bones. Unfractionated green fluorescent protein (GFP) expressing bone marrow cells (1–2 × 107) were transplanted into 10-week-old, male, lethally irradiated (total dose 9 Gy) wild-type C57BL/6 recipient mice by injection into the tail vein 5 h after irradiation. Four weeks after bone marrow transplantation, TAC or sham operation was performed as described earlier.16 Ramipril treatment started 3 days before surgery. For Tie-2-GFP bone marrow chimeras, Tie-2-GFP mice (FVB/NTgN[TIE2GFP]287Sato; Jackson Laboratory) were used as donors, and wild-type FVB/NJ mice (Charles River, Germany) were used as recipients.

2.3 Peripheral blood pressure

Peripheral blood pressure was obtained on the tail artery of mice 4 weeks after sham surgery or aortic constriction for 5 consecutive days using at least 20 repeating measurements per day (BP 2000 Series II Blood Pressure Analysis System, Visitech Systems, NC, USA).

2.4 Fluorescence-activated cell sorter analysis

Blood and bone marrow from 8–13 animals per group were analysed as described.10,16,18 The viable lymphocyte population was analysed for stem cell antigen 1 (Sca-1)-FITC (E13–161.7, Pharmingen, Germany) and vascular endothelial growth factor receptor 2 (Flk-1; Avas12α1, Pharmingen, Germany) conjugated with the corresponding phycoerythrin-labelled secondary antibody (Sigma, Germany). Isotype-identical antibodies served as controls (Becton Dickinson, Germany).

2.5 Culture of spleen-derived EPCs and migration assay

In mice, the spleen functions as a haematopoietic organ. Spleen mononuclear cells from at least six animals per group were isolated and cultured in endothelial basal medium (Cell Systems, Germany) as described.10,14,16,18 Antibiotics, calf serum, and cell culture medium were obtained from Invitrogen. After 4 days in culture, 500 cells from each mouse were transferred to modified Boyden chambers (BD Bioscience, Germany) in 24 well plates filled with 750 µL medium containing 10 µL SDF-1 (R&D Systems, Germany) to assess their migratory capacity. After 24 h of incubation, tetramethylindocarbocyanine-labelled acetylated low-density lipoprotein (DiLDL, 2.4 µg/mL; CellSystems, Germany) was added to identify EPCs. Boyden chamber filters were cut out, placed on slides, and mounted with fluorescent-mounting medium (Vectashield, Vector Laboratories, Burlingame, CA, USA) for fluorescence microscopic analysis. The whole filter was analysed using a Nikon E600 epifluorescence microscope (Nikon, Japan) with appropriate filters. Cells positive for red DiI-Ac-LDL were judged to be EPCs and were counted.

2.6 Disc angiogenesis model

A disc of polyvinyl alcohol sponge (Rippey, Germany), covered with nitrocellulose cell impermeable filters (Millipore, Germany), allowed capillaries to grow only through the rim of the disc. The discs were implanted subcutaneously. After 14 days, space-filling fluorescent microspheres (0.2 mm; Invitrogen, Germany) were injected into the LV to deliver them to the systemic microvasculature. The area of the disc invested by fibrovascular growth was assessed with Lucia Measurement version 4.6 software.

2.7 Immunofluorescence analysis

The protocols to detect cardiomyoyctes, endothelial cells, and GFP are described in detail in the Supplementary material online.

2.8 Apoptosis detection

To detect apoptosis, 6 µm thin paraffin sections of formalin-fixed mice heart tissue were examined using the ApopTag Peroxidase In Situ Oligo Ligation Kit from Millipore, according to manufacturer's instructions. The in situ oligo ligation assay specifically detects apoptosis by staining only cells that contain double-stranded breaks that are blunt-ended or have a one-base 3′ overhang (cells containing nicked, gapped, 3′-recessed, 3′-overhanging ends longer than one base, and single-stranded ends are not detected). Unlike conventional terminal transferase-based labelling, the assay stains apoptotic but not necrotic or transiently damaged cells.19 To distinguish whether the apoptotic cells were cardiomyocytes, endothelial cells, or other cardiac cells, specific immunostaining was performed after the apoptosis assay. Apoptosis was detected using light field microscopy for the brown 3–3′diaminobenzidine staining of the apoptosis kit, and the co-immunostaining for the respective marker was evaluated by switching to the fluorescence unit of the same microscope.

2.9 Tissue morphometry

For morphometric analyses, left ventricular tissue sections (3 µm) were examined. Cardiomyocyte short-axis diameter, degree of cardiac fibrosis, apoptotic indices, and the ratio of CD31pos cells to cardiomyocytes were evaluated. The procedures used for morphometric analyses were provided in detail elsewhere.16

2.10 Real-time polymerase chain reaction

Real-time quantitative reverse transcriptase–polymerase chain reaction (RT–PCR) was performed with the TaqMan system (AB Step One Plus, Applied Biosystems, Germany). The TaqMan probes and primers for mouse genes hypoxia-inducible factor-1α (Hif1α) (Mm00468869_m1) and 18S ribosomal RNA were assay-on-demand gene expression products purchased from Applied Biosystems. For quantification, mRNA expression of Hif1α was normalized to the expression of the housekeeping gene 18S ribosomal RNA.

2.11 Western blotting

Protein isolation and western blotting were performed as described in detail in Supplementary material online.

2.12 Statistical analysis

Results are presented as mean ± standard error of the mean. Unpaired t-test, Mann–Whitney test, or two-way ANOVA with a Bonferroni post hoc test was used where applicable. Values of P < 0.05 were considered significant.

3. Results

3.1 ACE inhibition prevents myocardial hypertrophy independent of blood pressure lowering

TAC [360 µm, 35 days, n = 29; sham group (SHAM) n = 26] increased left ventricular systolic pressure to 118 ± 6 mmHg in TAC vs. 78 ± 5 mmHg in sham-operated mice (P < 0.0001). This effect was not changed by treatment with ramipril 5 mg/kg p.o. (R-SHAM 71 ± 6 mmHg, n = 11; R-TAC 112 ± 6 mmHg, n = 25) (Figure 1A). Systolic blood pressure of the tail artery decreased significantly in mice after aortic ligation to 97 ± 2 mmHg compared with 113 ± 4 mmHg in sham-operated mice (P < 0.001). Ramipril in a dose of 5 mg per kg body weight did not significantly influence this parameter in both groups (R-SHAM 106 ± 4 mmHg; R-TAC 92 ± 2 mmHg) (Figure 1B).

Figure 1

Effect of ramipril on left ventricular and peripheral blood pressure and on cardiac and cardiomyocyte hypertrophy. Effect of ramipril, 5 mg/kg/day p.o. and transaortic constriction (TAC, 360 µm, 35 days) on left ventricular (LV) systolic pressure (A), on tail artery systolic pressure (B), on the ratio of heart weight to tibia length (C), and on cardiomyocyte short-axis diameter (D). n.s., not significant, *P < 0.05, #P < 0.01, §P < 0.001, **P < 0.0001, ***P < 0.000001.

TAC increased the ratio of heart weight to tibia length (mg/mm) from 8.4 ± 0.3 in sham-operated mice to 13 ± 0.9 after TAC (P < 0.001). Ramipril did not influence this ratio in the sham group (R-SHAM, 9.4 ± 0.5, n.s. vs. SHAM) but decreased it in the mice receiving TAC (R-TAC, 11 ± 0.6, P < 0.05 vs. TAC) (Figure 1C).

TAC for 35 days led to an increase in cardiomyocyte short-axis diameter (TAC 15.5 ± 0.4 µm) compared with sham-operated animals (SHAM 10.5 ± 0.4 µm, n = 8 per group, P < 0.000001 vs. TAC). Ramipril attenuated this effect (R-SHAM 10.6 ± 0.2 µm, n.s. vs. SHAM; R-TAC 12 ± 0.3 µm, P < 0.000001, R-TAC vs. TAC) (Figure 1D; see Supplementary material online, Figure S7).

TAC markedly increased cardiac fibrosis quantitated morphometrically as fractional area of collagen content in percentage of myocardial content using picrosirius red staining (SHAM 0.38 ± 0.12%, TAC 2.34 ± 0.86%; n = 6–12 per group, P < 0.05). This effect of TAC was abolished by ACE inhibition (R-SHAM 0.57 ± 0.1%, n.s. vs. SHAM; R-TAC 0.54 ± 0.35%, P < 0.05 vs. TAC) (Figure 2A; see Supplementary material online, Figure S8AD).

Figure 2

Effect of ramipril on cardiac fibrosis and apoptosis rates. (A) Effect of ramipril on cardiac fibrosis evaluated by collagen content (expressed as percentage of myocardial content). (BD) Influence of ramipril on cell-specific apoptosis rates of cardiomyocytes (B), non-cardiomyocytes (C), and endothelial cells (D). n.s., not significant, *P < 0.05, #P < 0.01, §P < 0.001.

In the animals examined on day 1 after TAC or sham surgery, the four groups (SHAM, TAC, R-SHAM, R-TAC) did not differ in heart weight to tibia length, cardiomyocyte diameter, or collagen content.

3.2 ACE inhibition prevents apoptosis in cardiomyocytes and endothelial cells but not in non-cardiomyocytes

TAC led to significantly enhanced apoptosis of cardiomyocytes (0.7 ± 0.2%) and of non-cardiomyocytes (1.0 ± 0.3%) in the heart compared with sham-operated mice (0.02 ± 0.01% and 0.12 ± 0.04%, respectively; n = 6–9 per group; P < 0.05 for both TAC vs. SHAM). The extent of cardiomyocyte apoptosis was potently decreased in TAC mice receiving ramipril (R-SHAM 0.04 ± 0.04%, n.s. vs. SHAM, R-TAC 0.09 ± 0.04%, P < 0.05 vs. TAC). In contrast, the percentage of apoptotic non-cardiomyocytes was not significantly influenced by ramipril (R-SHAM 0.05 ± 0.04%, R-TAC 1.1 ± 0.3%) (Figure 2B and C; see Supplementary material online, Figure S9A and B).

The percentages of apoptotic CD31pos cells increased significantly with elevated afterload (0.02 ± 0.01% in SHAM vs. 0.48 ± 0.23% in TAC, n = 5–8 per group, P < 0.001, TAC vs. SHAM). Ramipril treatment reduced significantly the levels of apoptotic CD31pos cells in TAC mice receiving ramipril. (R-SHAM: no apoptotic CD31pos cells, R-TAC 0.08 ± 0.06%, P < 0.05 R-TAC vs. TAC) (Figure 2D; see Supplementary material online, Figure S9C and D).

3.3 Early upregulation of EPC during treatment with ramipril

The effect of ramipril and increased cardiac afterload induced by TAC (360 µm, 1 day and 5 weeks after surgery) on the numbers of EPC in peripheral blood (n = 5–22 per group) and bone marrow (n = 5–13 per group) is depicted in Figure 3AD. Aortic constriction elevated EPC levels after 1 day compared with base levels in sham to 138 ± 10% in peripheral blood (P < 0.05) but not in bone marrow (113 ± 7%, n.s.). Ramipril treatment for 3 days prior to surgery increased EPC numbers in sham mice to 188 ± 27% (P < 0.05) and to 167 ± 20% (P < 0.01) of baseline in the blood and the bone marrow, respectively. In mice, after aortic constriction, short-term ramipril feeding raised the levels even to 201 ± 17% (P < 0.01) in the peripheral blood and to 145 ± 14% (P < 0.05) in the bone marrow. After 5 weeks, TAC led to the upregulation of EPC numbers to 145 ± 15% (P < 0.05) and to 164 ± 22% (P < 0.05) of baseline in the blood and the bone marrow, respectively. At this time point, EPC of ramipril-treated sham animals did not differ from baseline in the peripheral blood (R-SHAM 87 ± 11%), nor the bone marrow (110 ± 5). However, in animals after TAC, ramipril treatment decreased the EPC levels in the peripheral blood to 102 ± 6% (P < 0.05) and in the bone marrow to 109 ± 5% (P < 0.05) compared with the baseline levels of the sham animals.

Figure 3

Effect of ramipril on EPC numbers and Hif-1α expression. Effect of ramipril on EPC numbers in peripheral blood (A and B) and bone marrow (C and D) of mice after transaortic constriction (TAC) or sham operation for 1 day (A and C) or 35 days (B and D). EPC were measured by double labelling for stem cell antigen 1 (sca-1) and VEGF-receptor-2 by FACS analysis. Numbers are expressed as percentage of the EPC number measured in the relevant sham group. (E) TAC for 1 day leads to the upregulation of hypoxia-induced factor 1α (Hif-1α) in the heart, determined by real-time PCR. Additional ramipril treatment for 4 days to SHAM or TAC leads to the downregulation of Hif-1α m-RNA. (F) Thirty-five days after surgery, Hif-1α mRNA levels were downregulated in the TAC group as well as in both ramipril-treated groups compared with SHAM. (G and H) Protein expression of hypoxia-induced factor 1α (Hif-1α) normalized against β-tubulin. (G) TAC for 1 day leads to the upregulation of the protein level of Hif-1α in the heart, determined by western blotting. Additional ramipril treatment for 4 days to SHAM or TAC leads to the downregulation of Hif-1α protein. (H) Thirty-five days after surgery, Hif-1α protein levels were downregulated in the TAC group as well as in both ramipril-treated groups compared with SHAM. n.s., not significant, *P < 0.05, #P < 0.01.

3.4 Ramipril decreases Hif1α mRNA and protein expression

Real-time PCR of myocardial tissue revealed increased Hif1α mRNA expression after 1 day in animals after aortic ligation (SHAM 100 ± 20%, TAC 320 ± 109%; n = 5–8 per group, P < 0.05). mRNA levels of Hif1α were significantly decreased in animals receiving ramipril (R-SHAM 29 ± 6%, P < 0.05 vs. SHAM; R-TAC 30 ± 14%, P < 0.05 vs. TAC) (Figure 3E). Five weeks after the surgery procedure, Hif1α mRNA was downregulated not only in the ramipril-treated mice but also in the untreated ligated animals (n = 5–7 per group, SHAM 100 ± 47%, TAC 18 ± 9%, P < 0.05 vs. SHAM; R-SHAM 5 ± 2%; P < 0.05 vs. SHAM; R-TAC 10 ± 4%) (Figure 3F).

Similar to the mRNA expression, protein levels of myocardial Hif1α increased after 1 day in animals after aortic ligation (SHAM 100 ± 29%, TAC 301 ± 81%; n = 4–5 per group, P < 0.05) and were significantly decreased by ramipril (R-SHAM 9 ± 5%, P < 0.05 vs. SHAM; R-TAC 13 ± 3%, P < 0.01 vs. TAC) (Figure 3G). Five weeks after surgery, Hif1α protein was downregulated in the ligated mice and in both ramipril-treated mouse groups compared with SHAM (n = 4–5 per group, SHAM 100 ± 25%, TAC 21 ± 13% P < 0.05 vs. SHAM; R-SHAM 19 ± 9%; P < 0.05 vs. SHAM; R-TAC 27 ± 11%) (Figure 3H).

3.5 Ramipril enhances migratory properties of EPC

TAC for 1 day or 35 days did not have a significant effect on the number of endothelial cells migrating through the mesh of a Boyden chamber (0.3 cm2) compared with sham mice (1 day: n = 4–12 per group; TAC 99 ± 26 vs. SHAM 100 ± 30; 35 days: n = 8–11 per group; TAC 149 ± 33 vs. SHAM 130 ± 22). In contrast, ramipril treatment enhanced the number of migrating cells in both treatment groups after 1 day (R-SHAM 254 ± 18, P < 0.05 vs. SHAM; R-TAC 266 ± 41, P < 0.05 vs. TAC) (Figure 4A) and after 35 days (R-SHAM 252 ± 34, P < 0.01 vs. SHAM; R-TAC 239 ± 20, P < 0.05 vs. TAC) (Figure 4B).

Figure 4

Ramipril effects on EPC migration and neovascularization. Ramipril but not transaortic constriction (TAC) enhances the number of migrating EPC in a Boyden chamber with an insert size of 0.3 cm2 after 1 day (A) and after 35 days (B). Both ramipril and TAC enhance neoangiogenesis in a disc angiogenesis model after 5 weeks. Subcutaneous implantation of polyvinyl sponge for 5 weeks resulted in in-growth of new vessels. (C) Quantitative histomorphometrical measurements. Representative examples of vascularized areas around borders of discs in animals after sham-surgery (D), TAC (E), sham surgery and ramipril (F), and TAC with ramipril treatment (G), represented by bright green fluorescence of the injected microspheres in the newly formed capillaries. Bars: 1 mm. n.s., not significant, *P < 0.05, #P < 0.01.

3.6 Ramipril increases pressure overload-induced extracardiac and cardiac angiogenesis

TAC performed simultaneously with implantation of a subcutaneous polyvinyl disc (n = 5–17 per group) increased the area of neoangiogenesis by 55 ± 10% compared with sham-operated mice (P < 0.05). Oral treatment with ramipril (38 days) was able to even further enhance this effect in both groups (R-SHAM by 105 ± 15%, P < 0.05 vs. SHAM; R-TAC by 142 ± 2%, P < 0.05 vs. TAC) (Figure 4CG).

The ratio of CD31pos endothelial cells to α-sarcomeric actin-positive cardiomyocytes (n = 8–17 per group) was not changed 35 days after TAC (0.76 ± 0.03 in sham vs. 0.84 ± 0.08 in TAC). In contrast, feeding of ramipril 5 mg/kg increased this ratio significantly in both groups (R-SHAM 1.1 ± 0.05, P < 0.000001 vs. SHAM; R-TAC 1.1 ± 0.05, P < 0.01 vs. TAC) (Figure 5).

Figure 5

Effect of ramipril on cardiac capillary density. The effect of TAC on capillary density is expressed as number of CD31pos cells per number of cardiomyocytes (A). n.s., not significant, #P < 0.01, ***P < 0.000001. Cardiac sections from mice after sham procedure (B), TAC (C), sham procedure and ramipril (D), and TAC with ramipril treatment (E) with co-immunostaining for myocytic α-sarcomeric actin (green) and the endothelial cell marker CD31 (red). Nuclei are stained blue by DAPI. Bars: 20 µm.

3.7 Ramipril increases the extent of bone marrow-derived cardiac endothelial cells 5 weeks after TAC

Transplantation of GFP-positive bone marrow cells in lethally radiated wild-type mice (n = 8–19 per group) was used to determine whether ramipril has an influence on the integration and transdifferentiation of bone marrow-derived cells in the pressure-overloaded heart.

The influence of ramipril on left ventricular systolic pressure, blood pressure, heart weight/tibia length, cardiomyocyte short-axis diameter, and collagen content in these bone marrow-transplanted animals was similar to the parameters of animals which did not receive bone marrow transplantation (see Supplementary material online, Figures S10 and 11A). The extent of apoptosis in cardiomyocytes and non-cardiomyocytes, which was increased by TAC, was abolished by ramipril treatment (see Supplementary material online, Figure S11B and C). Apoptosis levels in endothelial cells were—as seen in the non-bone marrow-transplanted mice—elevated by increased afterload and were significantly reduced by treatment with the ACE inhibitor (see Supplementary material online, Figure S11D). These observations are described in the Results section of Supplementary material online in more detail.

To examine to what degree bone marrow-derived cells contribute to cardiac angiogenesis, co-staining of heart sections for CD31 and GFP was performed. In sham mice, 3.6 ± 1% of CD31pos cells were positive for GFP compared with 10.2 ± 1.5% post-TAC (P < 0.05). This effect was enhanced by ramipril in the pressure-overloaded mice but not in the sham group (R-SHAM 6.5 ± 2.7%, n.s. vs. SHAM; R-TAC 19 ± 4%, P < 0.05 vs. TAC) (Figure 6AE).

Figure 6

Influence of ramipril on the forming of endothelial cells from bone marrow-derived cells. The percentages of bone marrow-derived endothelial cells in the heart increased by transaortic constriction (TAC) after 5 weeks are further enhanced by ramipril treatment. (A) Quantitative measurements. n.s., not significant, *P < 0.05. (B) Staining for GFP (green), (C) CD31 staining (red), (D) nuclei stained blue by DAPI, (E) an overlay of the three stainings. A CD31-positive cell derived from the bone marrow, which is double-positive for GFP and CD31, is marked by arrow; endothelial cells not derived from the bone marrow (single-positive for CD31) are depicted by arrowheads. Bars: 20 µm. (F and G) Bone marrow-derived cells from Tie-2-GFP FVB/N mice migrate into the heart of wild-type FVB/N mice after bone marrow transplantation and transdifferentiate partly into endothelial cells as proved by green GFP expression driven by the endothelial cell promoter Tie-2. Endothelial cells are marked by arrows. Nuclei are stained blue by DAPI. Bars: 10 µm.

In agreement with our previous findings,16 we did not detect any large cardiomyocytes derived from bone marrow by co-staining with α-sarcomeric actin and GFP.

To assess whether cells from the bone marrow were transdifferentiating into endothelial cells in our model and to exclude the possibility that the findings were due to leucocytes or pericytes lying in the vicinity of true endothelial cells, transplantation studies of bone marrow cells from Tie-2-GFP FVB/N mice into their wild-type FVB/N mice were performed. All animals after bone marrow transplantation from Tie-2-GFP mice exhibited GFP-positive cells in the heart after surgery. GFP expression in these mice is under the control of the endothelial promoter Tie-2. In the bone marrow recipients, GPF-positive cells therefore represent endothelial cells of bone marrow origin (Figure 6F and G).

4. Discussion

Cardiomyocyte hypertrophy in pathological situations such as hypertension or valvular heart disease has been considered to be an adaptive response to increased external load, because hypertrophy can normalize the increase in wall stress induced by mechanical pressure overload. ACE inhibition in these pathological situations ameliorates cardiac and cardiomyocyte hypertrophy20,21 as demonstrated in our model of TAC. Inhibition of the ACE also abolished the fibrotic effects induced by pressure overload. This observation is in agreement with data from studies showing decreased induction of collagen I and III, laminin, and fibronectin,20 as well as decreasing the ratio of the gelatinolytic matrix metalloproteinase 2 to the tissue inhibitor of matrixmetalloproteinases 4 in the presence of ACE-I.22 Apoptosis of cardiomyocytes, endothelial cells, and non-cardiomyocytes increased significantly in mice after TAC.16 Interestingly, treatment with ramipril lowered the number of cardiomyocytes and endothelial cells undergoing apoptosis, but changes in the apoptosis of non-cardiomyocytes were observed in animals only after bone marrow transplantation. This amelioration of cell death of cardiomyocytes and endothelial cells is associated with an increased capillary–myocyte ratio in the presence of ramipril.

Heart size and cardiac function are angiogenesis dependent, and disruption of coordinated tissue growth and angiogenesis in the heart contributes to the progression from adaptive cardiac hypertrophy to heart failure in myocardial hypertrophy.15 In the literature, both pro-angiogenic5,23 and anti-angiogenic properties of ACE inhibitors have been described.23,24 The possible mechanisms for this phenomenon are still unclear since ACE inhibition also hampers neovascularization in other tissue.23 On the basis of our finding of upregulation of EPC during myocardial hypertrophy, we speculated that ACE inhibitors could exercise their positive role in the remodelling of myocardial hypertrophy through an effect on EPC.16 One day after induction of cardiac overload, animals receiving ramipril for 4 days showed an increase in EPC levels which was not detectable after 5 weeks. After induction of pressure overload for 5 weeks, EPC levels in the peripheral blood and in the bone marrow were increased in the vehicle but not in the ACE-I-treated mice. In the vehicle group, this may be the result of the insufficient capillary–myocyte ratio in the absence of an ACE-I which may persistently upregulate EPC numbers. Since ACE-I-treated animals had an improved capillary–myocyte ratio after 5 weeks but lower EPC levels, we speculate that after the amelioration of cardiac capillarization, a negative feedback mechanism leads to the downregulation of EPC. In our model of C57Bl6 mice, we did not observe a significant effect of 5 mg ramipril on peripheral blood pressure. Therefore, these effects of ramipril on EPC were not due to antihypertensive effects.

Ramipril treatment led to a significant amelioration of the migratory capacity of EPCs, whereas pressure overload did not alter the migratory properties of these cells. These data suggest that the functional improvement of EPCs significantly contributes to the improvement of cardiac and extracardiac angiogenesis in the ramipril-treated mice. Furthermore, short-time treatment with ramipril led to an upregulation of EPCs in the peripheral blood and in the bone marrow, whereas a short period of TAC alone led only to a rather modest increase of EPC levels in the peripheral blood but not in the bone marrow. Therefore, short time of pressure overload may have an influence only on the release of EPCs from the bone marrow but not on the production of EPCs, whereas the ACE inhibition influenced both the release and the production of EPCs.

Pharmacological interventions to increase the mobilization of EPC are not limited to ACE inhibition. Statin treatment has been shown to upregulate EPC in the systemic circulation14,25 which contributes to the improvement of cardiac function and increased capillary density in the setting of myocardial infarction.26 The effects of statins on EPC appear to be primarily mediated by endothelial nitric oxide; however, the effect of TAC on myocardial capillarization is not changed in eNOS−/− mice.16 Thum et al.26 found raised EPC levels after ACE inhibitor therapy in a model of myocardial infarction, which were independent of nitric oxide. Therefore, statins and ACE inhibitors may exert their effects on EPC via distinct signalling pathways, which opens the possibility for an additive effect of these two drugs. In patients with type 2 diabetes, treatment with an angiotensin receptor antagonist led also to increased numbers of EPC in the peripheral blood.27 Therefore, one could speculate that ACE inhibitors and angiotensin receptor antagonists may have a similar angiogenic potential using similar pathways. Clearly, further studies are needed to address these issues in detail.

Sano et al.28 showed that the Hif-1α induces angiogenic factors such as vascular endothelial growth factor during acute pressure overload; Hif-1α expression was rapidly and transiently upregulated during the first days after aortic ligation but no longer after a period of 28 days. We therefore examined the role of ramipril on the expression of Hif-1α in our model. We found that ramipril treatment for 4 days significantly reduced levels of Hif-1α m-RNA and protein in sham-operated animals as well as in mice after aortic ligation 1 day after surgery procedure. Thirty-five days after aortic ligation, Hif-1α m-RNA and protein were downregulated in the TAC group in agreement with the data of Sano et al.28 Interestingly, Hif-1α was downregulated compared with sham-operated, vehicle-treated control mice in all ramipril-treated groups. Therefore, the effects of ramipril on EPC in addition to the raise induced by pressure overload is independent of Hif-1α.

To examine whether EPC lead to a better capillarization, assessment of extracardiac neoangiogenesis and histological examination of the capillary–cardiomyocyte ratio were performed. After 5 weeks, extracardiac angiogenesis and cardiac capillary density were significantly enhanced in the ramipril-treated animals compared with the respective placebo-receiving groups. The experiments show that ramipril treatment and TAC have pro-angiogenic effects but while in pressure overload alone, the effect of angiogenesis is not sufficient; ramipril treatment in addition to pressure overload reverses the capillary–myocyte mismatch following apoptosis of both cell lineages. This could be in part due to the decrease of cardiomyocyte and endothelial cell apoptosis and in part due to an increase in newly formed endothelial cells stemming from the bone marrow. To further elucidate this issue, mice received GFPpos bone marrow transplantation after lethal irradiation to examine whether progenitor cells from the bone marrow would migrate into the heart and directly affect cardiac capillarization and whether the ACE inhibitor would influence these processes. In the animal groups undergoing bone marrow transplantation with GFPpos bone marrow cells, ramipril treatment led to a further increase of myocardial GFPpos bone marrow-derived CD31pos endothelial cells in the group with pressure overload.

The histological characterization of bone marrow cells migrating into the heart can be difficult because leucocytes or pericytes lay in the vicinity of endothelial cells.29 To overcome this issue, we used Tie-2-GFP mice as donor animals for additional bone marrow transplantation studies. In these animals, cells are expressing GFP driven by the endothelial Tie-2-promotor, and bone marrow-derived cells in this model only express GFP after differentiating towards endothelial phenotype. The data therefore strongly suggest that differentiation took place.

In summary, the main novel finding of the study is that the ACE inhibitor ramipril induces increased neoangiogenesis by the upregulation of EPC and amelioration of EPC migration during afterload-induced myocardial remodelling. The data show that EPC play a role during the development of cardiac hypertrophy. The regulation of progenitor cells from the bone marrow identifies a novel effect of ACE inhibitors in the treatment of pressure-induced hypertrophy to delay the maladaptive progression into ventricular dysfunction and heart failure.


This work was supported by the Alois Lauer Stiftung, Dillingen, Germany to P.M.; Universität des Saarlandes, Homburg, Germany (HOMFOR 06/46); Hans & Gertie Fischer-Stiftung, Essen, Germany to P.M.; Deutsche Forschungsgemeinschaft, Bonn, Germany (KFO 196) and the German Cardiac Society, Düsseldorf, Germany to A.K. The study was not supported by the pharmaceutical industry.


We thank E. Becker, S. Jäger, and I. Hartmann for their excellent technical assistance, and L. Kästner from the department of cell biology for his excellent support in using confocal microscopy. C. Roggia from the department of pathology supported us in the use of real-time polymerase chain reaction technology.

Conflict of interest: none declared.


  • Both authors contributed equally to this study and share first authorship.


View Abstract