If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Because both endothelin-1 (ET-1) and angiotensin II (AngII) are independent mediators of arterial remodeling, we sought to determine the role of ET receptor inhibition in AngII-accelerated atherosclerosis and aortic aneurysm formation. We administered saline or AngII and/or bosentan, an endothelin receptor antagonist (ERA) for 7, 14, or 28 days to 6-week- and 6-month-old apolipoprotein E-knockout mice. AngII treatment increased aortic atherosclerosis, which was reduced by ERA. ET-1 immunostaining was localized to macrophage-rich regions in aneurysmal vessels. ERA did not prevent AngII-induced aneurysm formation but instead may have increased aneurysm incidence. In AngII-treated animals with aneurysms, ERA had a profound effect on the non-aneurysmal thoracic aorta via increasing wall thickness, collagen/elastin ratio, wall stiffness, and viscous responses. These observations were confirmed in acute in vitro collagen sheet production models in which ERA inhibited AngII's dose-dependent effect on collagen type 1 α 1 (COL1A1) gene transcription. However, chronic treatment reduced matrix metalloproteinase 2 mRNA expression but enhanced COL3A1, tissue inhibitor of metalloproteinase 1 (TIMP-1), and TIMP-2 mRNA expressions. These data confirm a role for the ET system in AngII-accelerated atherosclerosis but suggest that ERA therapy is not protective against the formation of AngII-induced aneurysms and can paradoxically stimulate a chronic arterial matrix remodeling response.
Atherosclerosis and aneurysmal vascular diseases lead to substantial morbidity and mortality, and they share a number of clinical risk factors.
Both diseases are associated with inflammation, and, although some believe that advanced atherosclerosis may be a prerequisite for abdominal aortic aneurysm (AAA), not all patients with AAA have evidence of substantial atherosclerosis nor does atherosclerosis always lead to aneurysm formation. Indeed, several studies have suggested that aneurysms and atherosclerosis are influenced by distinctly different inflammatory processes with T helper (Th)1 cytokines predominating in atherosclerosis and Th2 responses predominating in aneurysms.
that suggest the utility of a common therapeutic approach toward each condition. However, it is also possible that potential therapeutic approaches may have divergent effects on both AAA and atherosclerosis.
Angiotensin II (AngII) has been associated with atherosclerosis and aneurysm rupture in humans
The AngII/apoE−/− mouse model is well characterized and represents the best model of combined atherosclerosis and aneurysm initiation; the striking features of this model are a marked infiltration of macrophages into the adventitia and media, the breakdown of elastin lamellae, and the development of marked dilation of the suprarenal aorta.
the incidence remains low but is markedly increased by AngII infusion. Pathologic differences may exist between young and old animals, and an appropriate comparison of the two under identical experimental conditions may help elucidate the importance of age and/or the presence of pre-existing lesions on AAA development.
Endothelin (ET)-1 is known to play an important role in atherogenesis,
yet its contribution to the formation of aneurysms is largely unknown. ET-1 is an important downstream mediator of many of the biological effects of AngII, having a pivotal role in vascular remodeling.
reported that increased arterial smooth muscle cell (SMC) hypertrophy in AngII-infused rats was completely inhibited with the administration of a selective ET(A)-receptor antagonist, LU135252. In this study, we have examined young and old apoE-null mice, confirming a modest contribution of ET-1 to atherogenesis in the AngII-infused model but also providing evidence of a previously unrecognized protective role for the ET-1 pathway in limiting the extent of fibrosis and remodeling of the aneurysmal matrix.
Materials and Methods
Animals and Design
Young (4 weeks) and old (6 months) apoE−/− mice (The Jackson Laboratory, Bar Harbor, ME) were examined for atherosclerotic progression and aneurysm formation after infusion with AngII with or without bosentan (Tracleer; Actelion Pharmaceuticals Ltd., Allschwil, Switzerland), a dual ET-1 receptor antagonist (ERA). Four-week-old mice (n = 180) were fed a Western-type diet (TD 88137; Harlan Teklad, Madison, WI) beginning 2 weeks before experimentation (mean cholesterol, 27.49 ± 1.63 mmol/L; mean triglyceride, 1.42 ± 0.17 mmol/L at sacrifice) to encourage lesion progression, and 6-month-old mice (n = 194) received normal chow. Invasive procedures were performed under anesthesia (xylazine 20 mg/kg and ketamine 100 mg/kg). All animal studies were conducted under protocols approved by the local animal care committee in accordance with guidelines from the Canadian Council of Animal Care.
In Vivo Model
Mice were randomly assigned to receive a 4-week infusion of either AngII (1000 ng · kg−1 · minute−1; A9525; Sigma-Aldrich, St. Louis, MO) or 0.9% NaCl by a subcutaneous osmotic minipump (Model 2004; Durect Corporation, Cupertino, CA). Mice assigned ERA received 10 mg · kg−1 · day−1 bosentan in their drinking water (a dose that significantly inhibited acute ET-1–induced rise in systemic blood pressure). After 4 weeks of treatment blood pressure was measured from the right common carotid artery with a fluid-filled catheter. Blood was extracted through cardiac puncture into EDTA vials, spun down (250 × g) and the plasma stored at −80°C. The arterial system was perfusion fixed with 4% paraformaldehyde at 70 mmHg pressure, and the aorta (arch to the iliac bifurcation) was removed, stripped of its adventitia, and photographed. Aneurysms were determined to be present if the widest region of the suprarenal aorta was 1.5 times greater than the upstream descending aorta. En face oil red O staining (O0625; Sigma-Aldrich)
was used to evaluate the percentage of luminal surface area occupied by lipid (atherosclerosis) in non-aneurysmal vessels with the use of ImageJ64 analysis software version 1.44 (National Institutes of Health, Bethesda, MD).
Serum total cholesterol and triglyceride concentrations were measured with an enzymatic cholesterol assay in a colorimetric procedure on a Technicon RA1000 analyzer (Bayer, Tarrytown, NY; n = 6). ET-1 levels from precipitated plasma samples after 2 and 4 weeks of treatment were analyzed through enzyme-linked immunosorbent assay (ELISA) with the use of commercially available kits [Endothelin (1–21) ELISA Immunoassay; ALPCO Diagnostics, Salem, NH] according to manufacturer's recommendations.
Tissue Mechanics Analysis
Aortas from a subset of the 6-month-old animals underwent uniaxial mechanical testing (n = 10 to 20). A 2-mm ring of the thoracic aorta was excised between the aortic arch and intercostal arteries and mounted between two orthodontic wire loops (0.4-mm diameter) to avoid gripping artifacts. The loops were mounted on a Mach-1 Micromechanical Test System (Biosyntech, Montreal, QC, Canada) to allow for the precise actuator-based circumferential extension of the tissue while load was recorded with a 1000-g load cell. During testing tissue was submersed in oxygenated Krebs solution (pH 7.4, and 37°C) containing 0.8 g/L papaverine (P3510; Sigma-Aldrich). A digital image of the tissue at 0.1-g resting tension was captured to determine segment (gauge) length and width. Stress was calculated from the wall thickness values measured from digital images of sections adjacent to the excised ring. Tissues were preconditioned at 20 cycles to 7-g load (approximately 15% of maximum load) with a strain rate of 10 mm/minute. For stress relaxation testing, tissues were loaded to 7 g, and the load decay was observed for 100 seconds. Finally, tissues were preconditioned as above before loading at 10 mm/minute to fracture.
Collagen and Elastin Quantification
All chemicals are from Sigma-Aldrich unless specified. Tissues were dried and digested in cyanogen bromide (50 mg/mL in 70% in formic acid), and the supernatant fluid, along with the washed pellet, was collected and separately hydrolyzed overnight at 110°C in 6N HCl. Total collagen was determined with the hydroxyproline assay modified from Huszar et al.
The supernatant fluid was reconstituted in collagen assay buffer (0.26 mol/L citric acid, 0.21 mol/L glacial acetic acid, 0.88 mol/L sodium acetate.3H2O, 0.85 mol/L sodium hydroxide; pH 6.0) and free hydroxyproline in the sample was analyzed by reacting with Chloramine T (0.5 mol/L in n-propanol; Fisher Scientific, Ottawa, ON, Canada) for 20 minutes, then with Ehrlich solution and perchloric acid (15 minutes at 65°C), and followed with absorbance reading at 550 nm against a hydroxyproline standard (pH 6). Total collagen was calculated by assuming a 12.7% hydroxyproline content. Elastin content was determined as total protein (Ninhydrin assay) in the cyanogen bromide–insoluble pellet with absorbance read at 570 nm against a hydrolyzed elastin standard. Elastin purity was confirmed in four samples through amino acid analysis by quadrupole time-of-flight mass spectrometry (Advanced Protein Technology Centre, Hospital for Sick Children, Toronto, ON, Canada).
Histologic and IHC Staining
Cross-sections spanning the length of the aorta were stained with H&E, Verhoeff's van Gieson, and picosirius red. Routine immunohistochemistry (IHC) was performed with the following primary antibodies: AngII (rabbit polyclonal, 1:1000; IHC7002; Peninsula Laboratories, Inc., San Carlos, CA), mouse ET-1 (rabbit polyclonal, 1:600; IHC6901; Peninsula Laboratories, Inc.), and mouse macrophage (rat polyclonal Mac-3, 1:10; 550292; BD Pharmingen, Franklin Lakes, NJ). A substitution with pre-immune serum in place of the primary and/or secondary antibodies was used as negative controls. Quantification of Mac-3 staining in intimal areas was performed with digital planimetry.
Apoptosis was detected with the DeadEnd Fluorometric TUNEL System (G3250; Promega, Madison, WI), and data were confirmed by immunostaining with fluorescein-conjugated cleaved caspase-3 (Asp175) antibody (1:100; 9667; Cell Signaling Technology, Danvers, MA). Slides were also stained with cyanine 3–conjugated monoclonal anti–α-smooth muscle actin (1:200; C6198; Sigma-Aldrich) and ToPro3 nuclear counterstain (1:5000; T3605; Invitrogen, Carlsbad, CA). Fluorescence intensity was quantified by a blinded observer, using a fixed gain and power setting on a Nikon (Melville, NY) fluorescent microscope equipped with an epi-illumination single-band emitter filter cassette for the separate illumination of green (543 nm), red (633 nm), and blue (488 nm) fluorescence.
SMC Culture Model of Collagen Sheet Production
Rat A10 SMC (CRL-1476; American Type Culture Collection, Manassas, VA) from passages 14 to 17 were cultured at an initial density of 10,000 cells/cm2 in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and penicillin/streptomycin (100 U/100 μg/mL). At confluence, cultures were fed daily with freshly prepared 50 μg/mL L-ascorbic acid (A7631; Sigma-Aldrich) to encourage collagen production for a total of 28 days. At 15 days of ascorbate supplementation flasks were randomized to receive supplementation with AngII (10−7 mol/L; Sigma-Aldrich), ERA (10−5 mol/L), both compounds, or vehicle (n = 9 to 10 separate experiments).
Quantitative Real-Time RT-PCR
Total RNA was extracted from collagen sheets with the use of the RNeasy MiniKit (74104; Qiagen, Valencia, CA) with the on-column DNase digestion step (RNase-Free DNase Set; 79254; Qiagen). Total RNA (1 μg) was reverse transcribed to first-strand cDNA with the use of the Omniscript RT Kit (205111; Qiagen). Primer sequences used for collagen type 3 α 1 (COL3A1),
are the same as previously described. Primer sets used for collagen type 1 α 1 (COL1A1) was 5′-AAGGTTCTCCTGGTGAAGCTG-3′ and 5′-ATCACACCAGCCTGTCCACGG-3′, MMP-2 was 5′-ACACTGGGACCTGTCACTCC-3′ and 5′-ACACGGCATCAATCTTTTCC-3′, and interferon (IFN)-γ was 5′-GCCCTCTCTGGCTGTTACTG-3′ and 5′-CCAAGAGGAGGCTCTTTCCT-3′. cDNA amplification for each gene of interest was monitored with 2× SYBR Green PCR Master Mix (4309155; Applied Biosystems, Carlsbad, CA) with the use of the ABI PRISM Sequence Detection System (model 7900HT SDS, Applied Biosystems). All results were normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) expression (TaqMan Rodent GAPDH Control Reagents kit; 4308313; Applied Biosystems). mRNA levels were calculated with the ΔΔCt formula: target gene expression/GAPDH expression = 2(ΔΔCt) and are reported in arbitrary units.
Rat A10 cells were treated with AngII, ERA, both compounds, or vehicle (n = 3 to 4 per group) for 24 hours before being serum starved overnight. Cells were then harvested with trypLE (12563029; Invitrogen), incubated in serum-free Dulbecco's modified Eagle's medium for 1 hour, and plated at 26,000 cells/cm2 on plates precoated with type I collagen (PICL24P05; Millipore, Bellerica, MA). Cells were incubated for 1.5 hours at 37°C, and nonadherent cells were removed by washing with PBS. Six to eight images per treatment group were taken at ×10 magnification under an inverted light microscope, and the number of adherent cells per field were counted.
Rat A10 SMCs pretreated for 24 hours with AngII, ERA, both compounds, or vehicle (n = 3 to 4) were harvested with 2 mmol/L EDTA (E9884; Sigma-Aldrich) and resuspended at 1000,000 cells/mL in staining buffer (1% bovine serum albumin, 0.1% sodium azide in PBS) containing anti-β1-integrin antibody (mouse monoclonal, 1:500; MAB1965; Millipore) or mouse IgG1 isotype control (human, 1:500; 11-632-C100; Axxora, San Diego, CA) for 1 hour. Cells were centrifuged (400 × g, 5 minutes), washed, and incubated with Alexa Fluor 488 (goat polyclonal, 1:1000; A11001; Invitrogen) for 1 hour in the dark. Cells were washed, resuspended in staining buffer, and analyzed with a Cell Lab Quanta SC system (Beckman Coulter, Mississauga, ON, Canada).
χ2 testing was used to determine the significance of aneurysm incidence. Quantitative results are expressed as mean ± SEM unless otherwise stated, and differences were determined by nonpaired t-test, or one-way analysis of variance with post hoc tests performed with Bonferroni's or Dunnet's multiple comparison tests when appropriate. For tissue mechanics, the nonparametric Kruskal-Wallis test was used. Analyses were performed with GraphPad Prism version 5 (GraphPad Software, Inc., San Diego, CA). P values < 0.05 were considered significant.
Characterization of Animal Model
Mean arterial blood pressure was not significantly different in any of the treatment groups of the same age when measured under anesthetic conditions. In vivo telemetric blood pressure measurements were also performed in a subset of 8-week-old mice with the use of implantable telemetry (DSI PhysioTel PA-C10; Data Sciences International, St. Paul, MN) by carotid catheter placement. Measurements were taken for 1 hour, twice daily for 7 days to establish a baseline control, then treated with AngII (n = 3) or combined AngII and ERA (n = 4) for 7 days. These animals showed a modest increase in blood pressure measured after 7 days of treatment compared with baseline (101.2 ± 2.75 mmHg versus 124.9 ± 5.71 mmHg; P < 0.05; n = 7 for control versus treatment); however, no differences in mean arterial blood pressure between AngII and AngII/ERA treatment groups were observed (Figure 1). Plasma cholesterol and triglyceride concentrations were not significantly different in any of the treatment groups of the same age. Plasma concentrations of ET-1 from AngII-treated young mice were not elevated after 2 weeks (10.98 ± 1.92 fmol/mL) but trended higher by week 4; however, this was not statistically significant over controls (young: 15.36 ± 3.38 fmol/mL versus 7.94 ± 1.98 fmol/mL, P = 0.10; old: 12.92 ± 2.97 fmol/mL versus 7.68 ± 2.20 fmol/mL, P = 0.25; n = 10 to 14 for AngII versus control); old mice treated with ERA did show a significant increase in plasma ET-1 (young: 13.27 ± 2.20 fmol/mL, P = 0.09, n = 12; old: 19.87 ± 1.85 fmol/mL, P < 0.01, n = 5 for ERA versus control). ET-1 immunostaining was seen mainly in the intima and media of aortas from AngII-treated animals, regions that often stained positive for the macrophage marker Mac-3 (Figure 2A). ERA was able to diminish ET-1 staining in aortic cross-sections from old AngII-treated animals (Figure 2A).
AngII-Induced Atherosclerosis Is Partially Mediated by ET-1
AngII-treated 6-week-old mice showed an eightfold (P < 0.001) increase in en face lesion area (Figure 2B), whereas 6-month-old mice had a threefold increase (P < 0.01) compared with age-matched controls (Figure 2C). Administration of ERA reduced AngII-induced lesion area in the young mice (P < 0.01) but not in the old animals, possibly because of the presence of pre-existing lesions in the aortas of older apoE−/− mice.
AngII Treatment Induces Aneurysm Development and Increases ET-1 Production and Macrophage Infiltration in the Suprarenal Aorta
Saline-treated apoE−/− mice did not develop AAA (Figure 3A). In all groups of AngII-infused mice, a subset exhibited histologic evidence of aneurysm formation in the suprarenal aorta, including thickened media and adventitial layers, and disruptions to the elastic lamina (Figure 3). To our surprise, ERA failed to protect against AngII-induced AAA formation. Both AngII and AngII/ERA groups produced a significant increased incidence of aneurysms compared with controls (Figure 3, A, C, and D). Although the incidence of AAA in the AngII/ERA groups were 1.5- and 1.3-fold higher in young and old mice than in mice treated only with AngII, this did not reach statistical significance in a χ2 analysis (young: 20.6% AngII versus 32.6% AngII/ERA; old: 37.7% AngII versus 50.7% AngII/ERA; Figure 3, C and D; P < 0.001). A power analysis (power of 80%, α = 0.05) showed that the test of incidence at this level of difference is relatively insensitive and would require unfeasible group sizes (>300). These observations were confirmed in a separate study that used LU409422 (10 mg · kg−1 · day−1; Abbott GmbH & Co. KG, Wiesbaden-Delkenheim, Germany), a mixed ERA, in which co-administration of AngII and LU409422 resulted in 36.4% of young (n = 33) and 44.7% of old (n = 38) apoE−/− mice developing AAA.
Although old mice had a trend for higher incidence of AAA, no distinct histopathologic differences were observed in the aneurysms from either age group. In aneurysmal regions, macrophage infiltration was evident throughout the vessel wall and was extensive within the intima (Figure 3, B and E). Regions of macrophage infiltration stained positive for AngII and ET-1. In vessels with aneurysmal dilatation, ET-1 immunostaining also extended to the media and adventitia layers of the thoracic aorta (Figure 3F). Changes in the cellular composition in the aorta of 6-month-old animals were examined with immunofluorescent staining after 1 and 2 weeks of treatment, before the aneurysms had become established. AngII increased SMCs in the media from both thoracic and suprarenal segments (Figure 4A); this hyperplastic effect was abrogated by ERA in the thoracic aorta. Apoptosis (TUNEL and cleaved caspase-3 staining; data not shown) was extremely low in the media of all vessels, and treatment with AngII/ERA had no measurable effect (Figure 4, B and C).
Collagen Content Is Modulated by the ET Pathway
Collagen and elastin content of the thoracic tissues was measured. Fibrillar collagen was increased in the thoracic aorta in response to AngII/ERA treatment (Figure 5A; P < 0.05), which was largely because of the profound increase seen in animals exhibiting suprarenal AAA (3.5-fold; Figure 5B). Elastin content remained unchanged in these vessels, resulting in marked elevation in the collagen/elastin ratio. Picosirius red–stained sections observed with circular polarized microscopy showed higher fibrillar collagen birefringence in the medial layer of thoracic aortas from the AngII/ERA-treated mice with suprarenal aneurysms, compared with the other groups (Figure 5C).
Biomechanical Properties of Aortas
Functional changes in aortic structure were examined in thoracic aortic segments from 6-month-old mice. AngII/ERA-treated mice exhibited increased wall thickness (0.25 ± 0.02 mm versus 0.15 ± 0.03 mm for ERA control; P < 0.05), cross-sectional area (0.73 ± 0.10 mm2 versus 0.37 ± 0.03 mm2 for control; P < 0.05), and vessel radius (0.50 ± 02 mm versus 0.43 ± 0.01 mm for control; P < 0.01) of the thoracic aorta. A subset analysis showed that these changes were greatest in thoracic aortas from AngII/ERA-treated animals with suprarenal aneurysms (wall thickness, 0.31 ± 0.04 mm; cross-sectional area, 1.01 ± 0.18 mm2; radius, 0.54 ± 0.03 mm).
Marked differences in stress-strain responses were observed between the groups (Figure 5D). The most pronounced biomechanical changes occurred in the thoracic aorta of AngII/ERA animals that developed suprarenal aneurysms. These changes were consistent with decreased elasticity and increased stiffness: elastic moduli (calculated at physiological stress levels) was substantially increased and stress-relaxation (percentage of stress remaining at 100 seconds) was reduced. Failure mechanics showed increased ultimate tensile stress and reduced yield strain in AngII/ERA animals with aneurysms (Table 1). These mechanical results are consistent with an increased fibrillar collagen content.
Table 1Mechanical Testing as Quantified from the Thoracic Rings of 6-Month-Old, Treated apoE−/− Aortas
Model validation was established through acute AngII dose-response studies that confirmed the up-regulation of COL1A1 mRNA with ≥10−7 mol/L AngII (expression at 24 hours was greater than 6 hours by 15-fold at10−6 mol/L and 43-fold at 10−5 mol/L).
Because matrix turnover (collagen synthesis and degradation) will influence the total collagen accumulation in the long-term treatment model, we compared the total collagen harvested with qualitative real-time RT-PCR for genes known to regulate collagen production. No significant increase in the percentage of total collagen (normalized to total protein) was found in this SMC sheet model after 2 weeks of AngII treatment (Figure 6A). Unlike the concentration-dependent up-regulation of COL1A1 mRNA during acute stimulation with ≥10−7 mol/L AngII, transcript levels were suppressed after 2 weeks of continuous AngII treatment (P < 0.05); a similar trend was found with COL3A1 mRNA expression (Figure 6B; P = 0.06). This may be due to a context-dependent transcriptional response that involves the negative feedback regulation of type I collagen.
Quantitative real-time RT-PCR examination for regulators of tissue proteolysis showed a trend toward increased transcription of MMP-2 (P = 0.05) and decreased TIMP-1 and TIMP-2 mRNA expression (Figure 6C; P = 0.1 and P < 0.05, respectively) in the AngII groups.
ERA Encourages a Fibrotic Response in the AngII-Stimulated Matrix
Continuous treatment with a combination of AngII and ERA for 2 weeks nearly doubled collagen content (Figure 6A). Treatment with ERA was able to overcome the repressive effects of AngII on the matrix by maintaining baseline levels of COL1A1 mRNA (Figure 6D; P < 0.01), enhancing COL3A1 mRNA expression (P < 0.005) and suppressing MMP-2 transcription (Figure 6E; P < 0.05). Blockade of ET receptors during AngII stimulation was found to markedly induce TIMP-1 and TIMP-2 mRNA expression (Figure 6E; P < 0.005). No changes in membrane type-1 MMP transcripts were found (data not shown).
AngII/ERA Reduces Cellular Adhesion to Type I Collagen, Mediated in Part by Decreased β1-Integrin Expression
Because cellular adhesion can regulate extracellular matrix synthesis, we looked at changes in adhesion to type I collagen after treatment with AngII, ERA, or AngII/ERA combined. ERA treatment alone compared with control cells showed a trend for decreased adhesion to type I collagen (64.62% ± 8.02% normalized to control; n = 4), and AngII/ERA cells displayed a significant decrease in adhesion compared with AngII treatment alone (108.1% ± 17.29% versus 50.88% ± 5.22% normalized to control; n = 3; P < 0.05; Figure 6F). Furthermore, specific analysis of major type I collagen receptor, β1-integrin, by flow cytometry showed a marked decrease in β1-integrin expression in cells treated with ERA alone (59% reduction) and cells given a combination of AngII and ERA (67% reduction) compared with controls.
ERA Abrogates AngII-Mediated Down-Regulation of IFN-γ
Expression of IFN-γ, a Th1 cytokine, was markedly reduced in groups treated with AngII (Figure 7A; P < 0.005). However, this down-regulation returned to baseline in the AngII/ERA group (Figure 7B; P < 0.0001).
In this study, we did not report a significant difference in blood pressure measurements between the AngII and AngII/ERA treatment groups. Although others have shown that ERAs may reduce AngII-mediated hypertension,
as well as ERA dosage, probably have varying effects on blood pressure. Notably, the ERA dose used in our work is based on a clinically relevant dose of 10 mg · kg−1 · day−1. Our studies showed that ERA significantly reduced the AngII-enhanced lesion progression in young mice and lesion complexity in older mice, suggesting that ET-1 is substantially involved in AngII-induced atherosclerotic progression. Others have reported a 31% reduction in lesion area of fat-fed young apoE−/− mice given the ET(A)-receptor antagonist LU135252 for 30 weeks
Thus, the causal role for atherosclerosis in the pathogenesis of AAA remains in question. In our study, ERA treatment markedly reduced atherosclerotic burden in the aorta but did not reduce AAA incidence in either old or very young animals; interestingly, ERA treatment increased biological markers of AAA progression in the AngII model. It is possible that blocking ET receptor binding only partially inhibits AngII-induced vascular inflammatory responses and in so doing promotes the development of a profibrotic phenotype. Recent studies suggest that conversion from Th1 to Th2 immune responses, including a deficiency in IFN-γ, are hallmarks of aneurysmal progression.
The marked plaque reduction with ERA seen in young mice suggests that, in the forming atheroma, ET-1 signaling is involved in AngII-mediated Th1 inflammation. However, AngII was shown to repress INF-γ expression in our cultured SMC sheet model which was restored by ERA, suggesting that AngII and ET-1 may have distinctly different effects in the vessel wall compared with T cells; this may in part account for the differential role of ET-1 in aneurysm and atherosclerosis. Although a marked inflammatory response was observed in all aneurysmal aortas, it was predominated by macrophages; in this situation ET-1 may be exerting its primary effect on fibrosis by modulating smooth muscle and macrophage responses.
Mechanical testing of proximal thoracic aortic segments from AAA aortas showed reduced distensibility and increased viscous properties compared with thoracic aortas from non-aneurysmal aortas and vehicle-treated controls; this response was most pronounced in AngII/ERA-treated animals with aneurysms. The mechanical findings are consistent with collagen accumulation and an elevated collagen-to-elastin ratio. Our results are consistent with the findings of Tham and colleagues
we found the transcription of both collagen types was only enhanced, in long-term culture studies, with the addition of an ET-receptor blocker. Thus, the effects of ERA on the vasculature are probably context dependent, promoting collagen production only in the presence of a chronic remodeling stimulus such as AngII.
The present experiments also confirmed the ability of AngII to induce MMP-2 expression and to reduce the secretion of its antagonist TIMP-2 in cultured SMC collagen sheets, a result previously reported in endothelial cells.
However, the combination of AngII and ERA decreased MMP-2 expression and increased expression of TIMP-1 and TIMP-2, factors that are conducive to fibrosis because TIMP-2 is known to inhibit active MMP-2 digestion of the formed matrix. Reports in the diabetic rat models found that inhibition of ET(B) markedly increased collagen deposition in mesenteric arteries, whereas ET(A) inhibition reduced collagen content and MMP-2 activity.
These studies reinforce the view that the role of ET-1 in vascular fibrosis is context dependent. In our experiments with ERA we found increased collagen deposition in AngII-treated groups only in the presence of established aneurysms or when SMCs were cultured in collagen sheets. Thus, the response to ERA that we observed may be contextually dependent on the presence or state of the collagen matrix.
Finally, our studies showed that combined treatment of AngII and ERA significantly reduced adhesion to type I collagen and that this may in part be due to decreased β1-integrin expression in the cells. α1β1- and α2β1-integrins are the main type I collagen receptors, and previous studies have shown that α1β1-integrins, which are highly expressed in vascular SMCs,
demonstrated loss or inhibition of α1β1-integrin induced enhanced collagen deposition in atherosclerotic mouse arteries. Our experiments suggest that AngII/ERA treatment might also abrogate this negative feedback regulation through decreased cellular adhesion to collagen and decreased β1-integrin expression, resulting in the up-regulation of collagen mRNA and overall increase in collagen synthesis and fibrotic response (summarized in Figure 8). This mechanism is consistent with our in vivo work and may help to explain the dramatic increase in collagen synthesis we observed in the aneurysmal aortas of apoE-null mice treated with AngII/ERA compared with those treated with AngII alone.
ERA treatment of cardiovascular disease has been controversial in the clinical setting, and therapeutic results seen in animal models (for a review, see Remuzzi et al
). Our study raises a possible concern with the suggested therapeutic use of ERA in patients with atherosclerosis. Although there may be a benefit in reducing the early progression of atherosclerosis, there may also be context-dependent effects on vascular matrix remodeling which could enhance fibrotic responses. Our model represents an early stage of aneurysm formation, an initial reparative process,
that involves marked inflammation, the production of cytokines that may result in widespread vascular fibrosis, focal elastin fragmentation, and weakening of the abdominal aortic wall that increases susceptibility to aneurysmal development. In this context it is possible that ET-1 plays a protective role by modulating the deleterious effects of chronic AngII infusion which may underlie the development of AAA. However, in the human setting, aneurysms develop over a much longer time course, and the precise role of ET-1 in all stages of this remodeling process is yet to be established.
We thank Dr. Catherine M. Bellingham, Dr. Yupu Deng, Dr. Fred W. Keeley, Dr. Effat Rezaei, Dr. Eva E. Sitarz, Dr. Qiuwang Zhang, and Dr. Yidan Zhao for their technical assistance.
Supported by the Canadian Institutes of Health Research (CIHR) operating grant MOP-74752 , the Pfizer/Canadian Hypertension Society/CIHR RxD Doctoral Research Award, and the Heart & Stroke/Richard Lewar Center of Excellence.
Bosentan was provided by Actelion Pharmaceuticals Ltd., Switzerland, and the LU409422 compound was provided by Abbott GmbH & Co. KG, Wiesbaden-Delkenheim, Germany.