|Home | About | Journals | Submit | Contact Us | Français|
The renin-angiotensin system (RAS) contributes to atherosclerotic lesion formation. Angiotensin converting enzyme 2 (ACE2) catabolizes angiotensin II (AngII) to angiotensin-(1–7) (Ang-(1–7)) to limit effects of the RAS. The purpose of this study was to define the role of ACE2 in atherosclerosis.
Male Ace2−/y mice in an LDL receptor deficient (Ldlr)−/− background were fed a high-fat (HF) diet for 3 months. ACE2 deficiency increased atherosclerotic area (Ace2+/y, 17 ± 1; Ace2−/y, 23 ± 2 mm2, P<0.002). This increase was blunted by losartan. To determine if leukocytic ACE2 influenced atherosclerosis, irradiated Ldlr−/− male mice were repopulated with bone marrow-derived cells from Ace2+/y or −/y mice and fed a HF diet for 3 months. ACE2 deficiency in bone marrow-derived cells increased atherosclerotic area (Ace2+/y, 1.6 ± 0.3; Ace2−/y, 2.8 ± 0.3 mm2; P<0.05). Macrophages from Ace2−/y mice exhibited increased AngII secretion and elevated expression of inflammatory cytokines. Conditioned media from MPMs of Ace2−/y mice increased monocyte adhesion to endothelial cells (HUVECs). Incubation of HUVECs with AngII promoted monocyte adhesion, which was blocked by Ang-(1–7). Co-infusion of Ang-(1–7) with AngII reduced atherosclerosis.
These results demonstrate that ACE2 deficiency in bone marrow-derived cells promotes atherosclerosis through regulation of AngII/Ang-(1–7) peptides.
Hypercholesterolemia stimulates components of the renin-angiotensin system (RAS), including angiotensin type 1 receptors (AT1R)1,2 and systemic concentrations of angiotensin peptides.3 Moreover, pharmacologic interference with the RAS through inhibition of ACE, renin or AT1R has been demonstrated to reduce both experimental and human atherosclerosis.4–6 Similarly, genetic manipulation of the RAS, including AT1aR3,7 or ACE deficiency8,9, caused a striking reduction in experimental atherosclerosis, further supporting a role for the RAS in atherosclerosis. Recent studies in our laboratories demonstrated that deficiency of renin in bone marrow-derived stem cells reduced hypercholesterolemia-induced atherosclerosis in LDL receptor (Ldlr−/−) deficient mice, suggesting that leukocyte production of angiotensin peptides influenced developing atherosclerotic lesions.6 Collectively, results support a prominent role of the RAS in developing atherosclerotic lesions; however, the relative importance of systemic versus local production of angiotensin peptides in lesion formation has not been fully resolved.
Angiotensin converting enzyme-2 (ACE2), a more recently discovered member of the RAS that has 40% homology to ACE10–12, is a monocarboxypeptidase that cleaves the vasoconstrictor, angiotensin II (AngII), to a vasodilatory peptide angiotensin 1–7 (Ang-(1–7)). ACE2 is located on the sex-linked X chromosome and exhibits widespread tissue distribution.10,13 While ACE2 can catabolize AngI, it has greater catalytic efficiency in the catabolism of AngII (~400-fold higher).12 Based on its ability to degrade the vasoconstrictor AngII and produce the vasodilator Ang-(1–7), ACE2 has been suggested to limit pathophysiologic activation of the RAS.
Previous studies localized ACE2 to atherosclerotic lesions of rabbit aorta and human carotid arteries.14,15 Interestingly, ACE2 localized to smooth muscle, endothelial cells, and macrophages in lesions of hypercholesterolemic rabbits.15 Functional studies have demonstrated that overexpression of ACE2 using adenoviral transfection in apolipoprotein E−/−(ApoE−/−) mice or in hypercholesterolemic rabbits decreased atherosclerosis.16,17 Recent studies demonstrated that whole body ACE2 deficiency increased atherosclerosis in ApoE−/− mice.18 In this study, we examined effects of whole body or leukocytic ACE2 deficiency on high fat (HF) diet-induced atherosclerosis in Ldlr−/− mice. To define mechanisms of ACE2 to regulate lesion formation, we focused on interactions between AngII and Ang-(1–7) as the substrate and product of ACE2, respectively, in the regulation of monocyte adhesion to endothelial cells and on lesion development.
To examine effects of whole body ACE2 deficiency on atherosclerosis, male Ace2+/y and Ace2−/y mice on an Ldlr−/− background (#002207; obtained from The Jackson Laboratories, Bar Harbor, MA) were fed a high fat diet (HF; 42% caloric intake from fat, TD88137, Harlan Teklad, Indianapolis, IN) for 3 months. In separate studies, losartan was administered by osmotic minipump (25 mg/kg/day, s.c.) to Ace2−/y male mice during month 3 of HF feeding. For bone marrow transplantation, male Ldlr−/− mice (2 months of age) were irradiated with a total of 900 Rads from a cesium source delivered in 2 doses.2 Mice were maintained on antibiotic water (sulfatrim, 4 ug/mL) for 1 week prior and 6 weeks after irradiation before feeding the HF diet for an additional 3 months. To define interactions between Ang-(1–7) and AngII in the regulation of lesion formation, Ang-(1–7) (400 ng/kg/min) was infused to fat-fed Ldlr−/− male mice for 7 days prior to co-infusion of AngII (1,000 ng/kg/min) for 28 days. Macrophages were harvested from the peritoneum of Ace2+/y or −/y Ldlr−/− mice. All studies were approved by the Institutional Animal Care and Use Committee at the University of Kentucky.
Deficiency of ACE2 (insertion of a neomycin cassette in exon 9 of the ACE2 gene19) was confirmed by ablation of mRNA of ACE2 in kidneys of Ace2+/− and −/− female mice (Figure 1A). Deficiency of ACE2 had no effect on body weight, systolic blood pressure, serum ACE activity, plasma renin concentrations, serum cholesterol concentrations or lipoprotein cholesterol distributions in Ldlr−/− mice fed a HF diet (Table 1, Supplemental Figure 2A). Plasma concentrations of AngII were modestly, but not significantly increased in Ace2−/y mice fed a HF diet (1 month of HF feeding: Ace2+/y, 29 ± 4; Ace2−/y, 39 ± 6 pg/mL; P>0.05). Whole body deficiency of ACE2 increased lesion area in aortic arches (Figure 1B & D, P<0.002) and aortic sinuses (Figure 1E, P<0.05) compared to Ace2+/y wild type controls. Quantification of CD68 immunostaining revealed that macrophage accumulation was greater in aortic sinus lesions from Ace2−/y mice compared to Ace2+/y Ldlr−/− controls (Ace2+/y, 0.25 ± 0.02; Ace2−/y, 0.44 ± 0.04 mm2; P<0.007)(Figure 1F). Administration of the AT1 receptor antagonist, losartan, to Ace2−/y mice decreased lesion formation in aortic arches (Figure 1C & D) and sinuses (Supplemental Figure 1A).
To define cell types implicated in effects of ACE2 deficiency, aortic sinus sections from Ace2+/y Ldlr−/− mice were immunostained for ACE2 or CD68 (Figure 2). Lesions with pronounced CD68 immunostaining also exhibited ACE2 immunoreactivity. In addition, ACE2 immunoreactivity was also evident in media and adventitia of aortic sinuses. Since ACE2 immunoreactivity localized to macrophages of lesions, we assessed ACE2 enzymatic activity in macrophages harvested from the peritoneal cavity (MPMs). ACE2 enzymatic activity in MPMs was similar to that in heart and liver (Supplemental Figure 1B).
Since lesional macrophages stained positive for ACE2, we repopulated irradiated Ldlr−/− mice with bone marrow-derived cells harvested from either Ace2+/y or Ace2−/y mice to define the role of leukocyte ACE2. After 6 weeks of repopulation, mice in each group were fed a HF diet for 3 months. Genomic analyses of irradiated mice demonstrated effective repopulation with Ace2+/y or Ace2−/y donor cells (Supplemental Figure 3). Deficiency of ACE2 in bone marrow-derived cells had no effect on body weight, systolic blood pressures, serum ACE activity, plasma renin concentrations, serum cholesterol concentrations, or lipoprotein cholesterol distributions (Table 1, Supplemental Figure 2B). ACE2 deficiency in bone marrow-derived cells increased lesion areas in both aortic arches (Ace2+/y, 1.6 ± 0.3; Ace2−/y, 2.8 ± 0.3 mm2; P<0.05, Figure 3A) and aortic sinuses (Figure 3B; P<0.05). In addition, CD68 positive immunostaining was increased in sections from aortic sinus of mice transplanted with Ace2−/y compared to Ace2+/y marrow (Figure 3C & D).
Concentrations of AngII released from MPMs of Ace2−/y mice were increased compared to Ace2+/y cells (Ace2+/y, 70 ± 8; Ace2−/y, 115 ± 13 pg/ml; P<0.05, Figure 4A). To determine if elevated AngII concentrations influenced inflammatory cytokines, we examined mRNA abundance and culture media protein levels of various inflammatory cytokines and/or their receptors using MPMs harvested from Ace2+/y or −/y Ldlr−/− mice. In MPMs from Ace2−/y mice, mRNA abundance of CCR2 was increased (Figure 4B), and protein concentrations of IL-6 and PAI-1 released into culture media were elevated (Figure 4C). In addition, deficiency of ACE2 resulted in modest elevations in mRNA abundance and/or protein release of CCL2, TNF-α and its receptor, and NF-kappa B (Supplemental Table 2 & 3). To determine if MPMs from Ace2−/y mice promoted monocyte adhesion to endothelial cells6, we co-cultured human vein umbilical endothelial cells (HUVECs) and MPMs from Ace2+/y and Ace2−/y mice by placing cells in the upper and lower chamber of Transwells, respectively. THP-1 monocytes were placed in the upper chamber to assay for adhesion to HUVECs. Co-culture of MPMs from Ace2−/y mice with HUVECs resulted in increased THP-1 monocyte adhesion compared to co-culture with MPMs from Ace2+/y controls (Ace2+/y, 100 ± 7; Ace2−/y, 142 ± 13% control; P<0.05; Figure 4D). Losartan reduced THP-1 monocyte adhesion when HUVECs were co-cultured with MPMs from Ace2+/y or −/y mice; however, losartan-mediated reductions in monocyte adhesion were more pronounced when HUVECs were co-cultured with MPMs from Ace2−/y (4-fold) compared to Ace2+/y mice (2-fold) (Figure 4D). In contrast, the Ang-(1–7) mas receptor antagonist, D-Ala, did not significantly influence THP-1 monocyte adhesion when HUVECs were co-cultured with MPMs from Ace2+/y or −/y mice (Figure 4D).
Recent studies demonstrated that Ang-(1–7) antagonized effects of AngII to promote vascular smooth muscle cell proliferation and migration.20 Therefore, we investigated the relative contribution of the ACE2 substrate, AngII, to the ACE2 product, Ang-(1–7), in the regulation of monocyte adhesion to HUVECs. Incubation of HUVECs with AngII increased monocyte adhesion, and this effect was abolished by losartan (Figure 5A). By itself, Ang-(1–7) had no effect on monocyte adhesion. However, Ang-(1–7) functionally antagonized AngII-induced stimulation of monocyte adhesion (Figure 5A), and this effect was blocked by D-Ala. To determine if these peptides interacted in vivo in the regulation of atherosclerosis, we infused Ldlr−/− mice with AngII in the absence or presence of co-infused Ang-(1–7). Co-infusion of Ang-(1–7) with AngII reduced (2-fold) atherosclerosis in the aortic arch (Figure 5B & E) and sinus (Figure 5D), and decreased CD68 immunostaining in aortic sinus sections (Figure 5C).
Results from this study demonstrate that whole body deficiency of ACE2 increased the development of atherosclerosis in fat-fed Ldlr−/− mice. ACE2 localized to macrophage-rich regions of atherosclerotic lesions, and ACE2 enzymatic activity was evident in mouse peritoneal macrophages. Interestingly, deficiency of ACE2 in bone marrow-derived cells also promoted the development of atherosclerosis. While plasma concentrations of AngII were not markedly elevated in whole body ACE2 deficient mice, macrophages from ACE2 deficient mice released greater concentrations of AngII. Moreover, macrophages from ACE2 deficient mice exhibited increased expression and release of inflammatory cytokines, and promoted monocyte adhesion to endothelial cells. While Ang-(1–7) had no effect on monocyte adhesion by itself, this angiotensin peptide functionally antagonized AngII-induced stimulation of monocyte adhesion and lesion formation. These results demonstrate that ACE2 deficiency promotes atherosclerosis, and suggest that endogenous ACE2 protects against atherosclerosis by controlling the relative balance between AngII and Ang-(1–7) in pivotal cell types.
We found that cultured macrophages from ACE2 deficient mice release greater concentrations of AngII. In contrast, whole body deficiency of ACE2 had no effect on systemic concentrations of renin or AngII. Our results are in agreement with several studies that document no change in plasma or kidney angiotensin peptide concentrations in ACE2 deficient mice.21,22 However, it should be noted that a previous study reported that ACE2 deficiency increased systemic concentrations of AngII in mice experiencing heart failure.13 An interesting aspect of the present study is that whole body and bone marrow cell deficiency of ACE2 promoted atherosclerosis, but had no effect on the systemic RAS. These results suggest that ACE2 primarily influences concentrations of AngII in pivotal cell types involved in developing lesions.
Previous investigators demonstrated that whole body ACE2 deficiency had strain-dependent effects on blood pressure.19 In C57BL/6 mice, ACE2 deficiency resulted in a modest increase in blood pressure (~7 mm Hg).19 We did not observe an effect of ACE2 deficiency on systolic blood pressures in Ldlr−/− mice. It is possible that ACE2 deficiency on an Ldlr−/− background blunted blood pressure increases in C57BL/6 mice. Previous studies demonstrated that although AT1a receptor deficiency had divergent effects on systolic blood pressure in Ldlr−/− and ApoE−/− mice, loss of AT1a receptor signaling resulted in pronounced reductions in atherosclerosis.3,7 Therefore, it is unlikely that blood pressure effects from ACE2 deficiency contributed to augmented atherosclerosis observed in the present study.
Our results are in agreement with previous studies demonstrating that ACE2 localized to macrophage-rich areas of atherosclerotic lesions from hypercholesterolemic rabbits.15 However, these results extend previous findings by demonstrating that murine macrophages exhibit ACE2 enzymatic activity and that macrophages from ACE2 deficient mice release greater concentrations of AngII. Similar to previous findings demonstrating reductions in thoracic aortic lesion areas following bone marrow transplantation in high fat fed Ldlr−/− mice6,23, in this study bone marrow transplantation reduced lesion areas in aortic arches of high fat-fed Ldlr−/− mice of both genotypes. Even though background levels of atherosclerosis were reduced in bone marrow transplanted mice, chimeric Ldlr−/− mice with ACE2 deficiency in bone marrow-derived cells exhibited increased atherosclerosis of a similar magnitude (~2-fold) as seen from whole body ACE2 deficiency. Moreover, chimeric mice with ACE2 deficiency in bone marrow-derived cells exhibited increased atherosclerosis in the absence of changes in the systemic RAS. These results suggest that local effects of ACE2 on bone marrow-derived cells, potentially macrophages, mediate effects of ACE2 deficiency to promote atherosclerosis.
Previous investigators demonstrated that adenoviral overexpression of murine ACE2 in rabbits subjected to endothelial injury and fed an atherogenic diet attenuated lesion progression.16 Similarly, recent studies demonstrated that adenoviral over-expression of ACE2 in atherosclerotic lesions of rabbits attenuated fatty streak formation.20 In ApoE−/− mice, adenoviral overexpression of ACE2 reduced atherosclerosis.17 Moreover, recent studies demonstrated that whole body deficiency of ACE2 in ApoE−/− mice increased atherosclerosis.18 In agreement, results from this study demonstrate that whole body deficiency of ACE2 in Ldlr−/− mice increased atherosclerosis. Collectively, these results support a pivotal role for ACE2 in developing atherosclerotic lesions.
Our studies extend previous findings by contrasting effects of whole body versus bone marrow deficiency of ACE2 on atherosclerosis. Results demonstrate a similar magnitude increase in lesion formation in whole body ACE2 deficient mice compared to chimeric mice lacking ACE2 in bone marrow derived cells. While these findings do not indicate a specific cell type responsible for augmented lesion formation in mice lacking ACE2 in bone marrow derived cells, macrophage rich regions of lesions from Ldlr−/− mice stained positively for ACE2. In addition, MPMs exhibited ACE2 enzymatic activity, and macrophages from ACE2 deficient mice released greater concentrations of AngII. Recent studies demonstrated that aortas from ACE2 deficient ApoE−/− mice exhibited increased cell adhesion when perfused with whole blood.18 Moreover, bone marrow derived macrophages from ACE2 deficient mice exhibited increased expression of inflammatory cytokines in response to LPS.18 Results from the present study demonstrate increased monocyte adhesion when HUVECs were co-cultured with MPMs from ACE2 deficient mice, and extend previous findings by demonstrating that these effects are AngII/AT1 receptor mediated.2 Moreover, in this study, in the absence of external stimuli (e.g., LPS)18, ACE2 deficient MPMs exhibited increased expression and release of inflammatory cytokines. These results demonstrate that deficiency of ACE2 in pivotal cell types promotes inflammation and monocyte adhesion, two mechanisms invoked in lesion formation.
In the setting of ACE2 deficiency where AngII is formed but not catabolized to Ang-(1–7), our results using losartan support a prominent role for AngII effects at AT1 receptors to augment atherosclerosis. In contrast, in wild type mice both AngII and Ang-(1–7) are present to regulate lesion formation. Recent studies demonstrated that long-term infusion of Ang-(1–7) to ApoE−/− mice reduced atherosclerosis.24 Moreover, recent studies demonstrated functional antagonism of AngII-mediated regulation of smooth muscle cell proliferation and migration by Ang-(1–7), supporting functional interactions between these angiotensin peptides that are regulated by ACE2.20 Similar to findings in smooth muscle cells, in the present study while Ang-(1–7) had no effect on monocyte adhesion when incubated alone with HUVECs, this peptide functionally antagonized effects of AngII to promote monocyte adhesion. Since effects of Ang-(1–7) to blunt AngII-induced monocyte adhesion were abolished by D-Ala, these findings implicate the mas receptor as the mediator of Ang-(1–7)-induced antagonism of AngII. Infusion of Ang-(1–7) with AngII markedly lowered atherosclerosis in Ldlr−/− mice, supporting an in vivo interplay between these ACE2-regulated angiotensin peptides in the development of atherosclerosis.
In conclusion, the present study demonstrated that whole body deficiency of endogenous ACE2, as well as deficiency in bone marrow-derived cells, increased atherosclerosis in Ldlr−/− mice. Elevated macrophage concentrations of AngII in ACE2 deficient mice promoted expression of inflammatory cytokines and increased monocyte adhesion to endothelial cells. When both AngII and Ang-(1–7) were present, Ang-(1–7) functionally antagonized effects of AngII to promote monocyte adhesion and lesion development. These results demonstrate that regulation of local concentrations of AngII versus Ang-(1–7) in pivotal cell types by enzymes such as ACE2 influence developing atherosclerotic lesions.
a) We acknowledge the excellent technical assistance of Victoria English, Debra Rateri and Michael Karounos. b) This work was supported by grants from the National Institutes of Health (F32 HL095281, SET and R01 HL062846, AD/LC) and by an Alaska Kidney Foundation Research Grant (SG).
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
c) Authors have no disclosures.