Ursodeoxycholic Acid Attenuates Acute Aortic Dissection Formation in Angiotensin II-Infused Apolipoprotein E-Deficient Mice Associated with Reduced ROS and Increased Nrf2 Levels Open Access

Acute aortic dissection (AAD) is a medical emergency, which is correlated with high mortality [1]. Treatment of this fatal vascular disease needs to be achieved by timely interventional therapy and surgical repair, but medical treatment has not yet achieved good clinical results. The fundamental reason for a lack of effective treatment is that the underlying pathological mechanisms responsible for aortic dissection remain elusive.

The roles of inflammation in the pathogenesis of aortic dissection and formation of aneurysms are well established [2,3]. Moreover, many studies have indicated that reactive oxygen species (ROS) also play an important role in the occurrence and development of aortic dissection. ROS are by-products derived from mitochondrial energy metabolism, NADPH, xanthine oxidase as well as from uncoupled endothelial nitric oxide synthase. Intracellular ROS are obliterated by superoxide dismutase (SOD), catalase and glutathione peroxidase. Redundant oxidative stress shifts the subtle balance between ROS generation and ROS elimination towards predominant ROS generation. Increasing evidence has suggested that NADPH oxidase is a source of oxidative stress in the pathogenesis of aortic dissection and formation of aneurysms both in murine disease and human disease [4,5]. ROS might affect the expression of matrix metalloproteinases and induce smooth muscle cells apoptosis, which are the predominant pathological alterations in diseases such as aortic dissection and aneurysms.

Ang II perfusion in apolipoprotein E-deficient (ApoE^-/-) mice has become a particularly useful animal model for investigating the pathological process of AAD. The formation of aortic dissection is similar to abdominal aortic aneurysm. The mechanism of the Ang II infusion model is closely associated with oxidative stress [6], and Ang II can also promote activation of NADPH oxidase and ROS production in vascular cells [7] and mononuclear cells [8].

UDCA possesses antioxidant properties, which include scavenging superoxide anions and inhibiting lipid peroxidation [9,10]. Nrf2, which is sequestered in the cytosol by Kelch-like Ech-associated protein (Keap1), acted as an important anti-oxidative molecule, was regulated by UDCA [11]. Evidences also show that UDCA plays a unique role in modulating the apoptotic threshold in both hepatic and non-hepatic cells [12,13,14]. Therefore, we investigated whether the antiapoptotic action of UDCA protects against formation of AAD via regulation of ROS.

ApoE^-/- mice (C57BL/6 background) were housed at the animal care facility of Tongji Medical College under specific pathogen-free conditions and fed a normal diet. The animal studies were approved by the Institutional Animal Research Committee of Tongji Medical College. Eight month-old male ApoE^-/-mice were randomly divided in saline group (n = 30), Ang II group (n = 37) and Ang II+UDCA group (n = 37), respectively. Mice were infused via osmotic mini pumps (Alzet, Cupertino, CA) with either saline or 2500 ng· kg^-1· min^-1 Ang II (Sigma-Aldrich, St. Louis, MO) for 7 days. UDCA (10 mg/kg/day in 0.1 mL 2.5% NaHCO₃, PH7.4) was administered via intragastric gavage for 3 consecutive days before Ang II infusion and also during the Ang II infusion for another consecutive 7 days (twice per day). Blood pressure was measured using the tail-cuff method described previously [15] and after implantation, and prior to sacrifice. The maximal aortic diameter was measured using the microscopy with NIS-Elements D.3.10 software attached to the microscope (Nikon, Tokyo, Japan). The aorta was fixed in situ at the end of 7 days Ang II infusion, immediately after animal sacrifice.

Aortic tissues from ApoE-deficient mice were harvested directly after completion of 7 days Ang II infusion. About 10 mice aortic tissues were obtained and kept frozen in liquid nitrogen and then stored at -80°C for western blot analysis and prepared for other experiments. Other aortic tissues were harvested, fixed in 4% paraformaldehyde in PBS, and embedded in paraffin for histological analysis. 4 µm cross-sections were prepared and subsequently stained with hematoxylin and eosin (H&E), van Gieson, TUNEL, DHE and F4/80, respectively. Immunohistochemical staining was performed according to the manufacture's description (Boster, Wuhan, China). F4/80 positive macrophages were counted in each evaluated micrograph field. Subsequently, the number of F4/80 positive macrophages per 1 mm² area was obtained by dividing the number of F4/80 positive macrophages by micrograph field area. Five to seven random micrograph fields were analyzed per group. The aortic dissection was defined histologically (H&E) as a splitting of the middle layer (media) from the outer layer of the aorta and/or blood accumulation within the aortic wall. The analysis was performed blinded by an experienced staff member of our histology center facility. Additionally, mice that were found dead without any preceding signs of suffering were defined as sudden death. Signs of suffering in mice were actively searched for by the following criteria: weight, activity, and fur appearance (during the course of the study, no sign of suffering in any of the mice was detected).

Mouse vascular smooth muscle cells (VSMCs) (ATCC, Manassas, VA) were cultured in 10% fetal bovine serum (Gibco, Grand Island, NY) containing Dulbecco's modification of Eagle's medium (Gibco, Grand Island, NY) under 37°C and 5% CO₂ conditions. After confluence, VSMCs were firstly pretreated with 300 µM UDCA (Sigma-Aldrich, St. Louis, MO), then incubated with 10 µM Ang II (Sigma-Aldrich, St. Louis, MO) and 0.5 mM H₂O₂ (Sigma-Aldrich, St. Louis, MO) for 24 h. For treatment, 5mM N-acetyl cysteine (NAC) (Sigma-Aldrich, St. Louis, MO) was also added 1 h before Ang II treatment.

Transfection was performed with Lipofectamine 2000 reagent (Invitrogen, CA, USA) according to the manufacturer's instructions. VSMCs were plated in 6-well plate at a density of approximately 1.2 × 10⁶ cells/ dish. The cells were harvested 48h after transfection with Nrf2-siRNA or Random-siRNA, which were chemically synthesized (Ruibo, Guangzhou, China).

The whole thoracic aortas from ApoE-deficient mice were first homogenized with polytron PT 1200E (Kinematica AG). SOD activity, lipid peroxidation, and catalase activity in aortic media tissues, serum and VSMCs were determined. SOD activity in aortic media homogenates was assayed based on a published method that measures the inhibitory rate of xanthine oxidase-mediated reduction of cytochrome c (pH 7.4). Lipid peroxidation was assayed by measuring malondialdehyde (MDA) levels that reacted with thiobarbituric acid at 535nm. Catalase activity was determined by monitoring the break-down of hydrogen peroxide catalyzed by catalase. Intracellular ROS production of VSMCs was assayed by flow cytometry (BD Biosciences).

The apoptosis and Necrosis Assay Kit with annexinV-FITC and PI were obtained from Beyotime (Beyotime Institute of Biotechnology, Shanghai, China). Briefly, the cultured VSMCs were trypsinized and then washed with PBS twice and suspended in 100 µl of cell staining buffer. A volume of 5 µl of annexinV-FITC and 5 µl of 10 µg/mL PI were added and incubated at 4°C or the cells were kept on ice for 20 min in the dark. After incubation, the apoptotic cells were immediately analyzed using flow cytometry (BD Biosciences). Normal cells showed red fluorescence, while apoptotic cells showed green fluorescence with JC-1 staining. TUNEL staining (terminal deoxynucleotidyl transferase dUTP nick end labeling) was performed according to manufacturer's instructions (In situ cell death detection kit, Fluorescein, Roche) to identify apoptotic cells. Total numbers of nuclei were identified by propidium iodide. The ratio of TUNEL-positive and total nuclei was calculated in ten high-power fields per thoracicaortic section.

Tissue samples and cells were harvested, homogenized and subsequently performed for western blot as previously described [16]. The whole thoracic aortas from ApoE-deficient mice were first homogenized with polytron PT 1200E (Kinematica AG). In vitro protein was isolated from VSMCs using a lysis buffer containing NP-40 (0.5%), PMSF, phosphatase and protease inhibitors. Protein concentration was determined by the Bicinchoninic Acid (BCA) kit (Sigma-Aldrich, St. Louis, MO) according to the routine procedures described previously [17].The following antibodies were applied: gp91, p67, p47, catalase, Cu/Zn-Sod, Ec-Sod, Bcl-2, Bcl-xl and Bax were from Epitomic (Burlingame, CA, USA); MMP2 and β-actin (Santa Cruz Biotechnologies, CA, USA). β-actin was used as calibration for total protein or cytosolic protein determination. Bands were quantified by densitometry using Quantity One software (Bio-Rad, Hercules, CA).

Data are reported as means ± SEM. All data analysis was performed with the use of SPSS 13.0 statistical software. The χ² test was performed to compare the AAD incidence between Ang II and Ang II+ UDCA group. In other experiments, one-way ANOVA or the Student's t test was used to determine statistical significance with P < 0.05. Each experiment was done at least in triplicate.

The protective effects of UDCA against Ang II infusion in ApoE^-/-mice were determined by both morphologic and histological method. Firstly, Ang II induced thrombosis formation was observed in isolated aortas, which was attenuated by UDCA pretreatment (Fig. 1A). Hematoxylin and Eosin (H&E) and Verhoeff's Van Gieson (EVG) staining confirmed that true and false lumens existed after Ang II infusion (Fig. 1B), a typical pathological character of AAD formation [18]. In addition, Ang II-infusion resulted in higher incidence of AAD (35%, 13/37) (Fig. 1C) and increased maximal aortic diameters measured at the suprarenal region of the abdominal aorta (1.42 ± 0.07 mm) compared with saline-infused control ApoE^-/-mice (0.96 ± 0.06mm). Interestingly, UDCA pretreatment markedly lowered the incidence of AAD formation (16%, 6/37) (Fig. 1C) and reduced the maximal aortic diameter (1.18 ± 0.33 mm) than Ang II infused mice (Fig. 1D). Finally, UDCA administration reduced Ang II induced mortality in the development of AAD (Fig. 1E). Together, UDCA pretreatment suppressed Ang II-induced AAD formation in ApoE^-/- mice.

Oxidative stress plays an important role in the pathogenesis of many vascular diseases [19,20] including AAD [21]. To examine whether UDCA exerts an antioxidant effect in the abdominal aorta during formation of AAD, dihydroethidium (DHE) staining was performed to detect the location of superoxide production within aortic tissues. Strong red fluorescence was observed in aortic tissue from Ang II-induced AAD mice, whereas the UDCA+Ang II group showed low-intensity of fluorescence, regardless with AAD formation or not (Fig. 2A). Moreover, the ROS production was equally enriched in the adventitia or in vascular smooth muscle cells (VSMCs), which was identified by α-SMA staining (Fig. 2B). And UDCA treatment equally prevented Ang II-induced ROS production both in adventitia and VSMCs, indicating the functions of UDCA on ROS reduction between adventitia and VSMCs have no specificity. Immunohistochemical staining (Fig. 2C) and western blotting (Fig. 2E) showed that administration of UDCA markedly inhibited aortic expression of the ROS-associated NADPH subunits gp91, p67 and p47, which were substantially activated by Ang II infusion.

Lipid peroxidation was determined by monitoring MDA levels, which were increased by administration of Ang II (Fig. 2D). SOD is an important antioxidative enzyme in the vessel wall [22]. Ang II decreased the activity of SOD and catalase. The expression of Cu/Zn-SOD, Mn-SOD and CAT was significantly lower in the Ang II group compared with the saline group, whereas these alterations were attenuated by UDCA pretreatment. Together, UDCA pretreatment prevented Ang II induced aortic oxidative stress productions.

Apoptosis of the aortic medial wall and adventitial cells played central roles throughout the progression of AAD. TUNEL staining was performed to detect apoptosis in aortic samples from mice infused with Ang II for 7 days. Immunofluorescence staining showed that Ang II produced overt apoptosis in the aortic media and adventitia (Fig. 3A), but UDCA pretreatment attenuated Ang II induced apoptosis both in VSMCs and in adventitia, but more obvious in VSMCs. We speculated that perhaps UDCA prevented AAD mainly focusing on the medial cells. Western blotting showed that Ang II infusion led to significantly lower Bcl-2 and Bcl-xl levels, and higher Bax levels compared with the saline group. Interestingly, UDCA pretreatment prevented Ang II induced alterations of Bcl-2, Bcl-xl and Bax (Fig. 3B).

Additionally, aortas from Ang II-infused mice showed induction of infiltration of macrophages (marked by the presence of F4/80) (Fig. 3C), which was more evident in adventitia. In contrast, UDCA mainly inhibited the infiltration of macrophages in adventitia after Ang II-infusion. Meanwhile, the alterative levels of inflammatory factor TNF-α, IL-10, IL-1β and IFN-γ in serum samples were in consistence with infiltration of macrophages (Fig. 3D).

Immunohistological staining and western blotting showed that UDCA resulted in a lower expression of MMP2 and MMP9 in aortas, which was increased in Ang II-infused ApoE^-/- mice (Fig. 4A, B), indicating UDCA inhibited the expression of MMP2 and MMP9 in the attenuation of AAD's formation. In vitro experiments in VSMCs confirmed that UDCA reduced the expression of MMP2 by affecting ROS (Fig. 4C, D).

Together, UDCA attenuates Ang II induced apoptosis of aortic VSMCs and adventitia, mainly focused on the reduction of medical layer apoptosis. Additionally, UDCA attenuates infiltration of macrophages mainly in adventitia.

To confirm the anti-oxidative feature of UDCA, the ROS production in VSMCs was measured after different stimulations. Obviously, Ang II-induced increase of ROS was significantly attenuated by pretreatment of UDCA, the effect of which was comparable with pretreatment with NAC, an effective ROS scavenger (Fig. 5A) [23]. Additionally, the ROS production induced by challenge of Ang II was comparable to the result of ROS stimulator H₂O₂ (Fig. 5A), indicating UDCA possessed a potent anti-oxidative property.

Previous studies showed that NF-E2-related factor 2 (Nrf2) was a potent anti-oxidative molecule and UDCA increased the expression of Nrf2. We next investigated whether UDCA exhibited anti-oxidative property via Nrf2. Western blotting showed that UDCA affected the expression of Nrf2 in a time and dose dependent manner in VSMCs, which exerted most significant effects at 24h by using the concentration of 300 µM UDCA (Fig. 5B-C). The ROS-associated NADPH subunits gp91, p67 and p47 were markedly increased when stimulated with Ang II, the effects of which were reversed by UDCA. In the presence of Nrf2-siRNA, the reduced effects of gp91, p67 and p47 were not observed (Fig. 5D). In addition, expression of the anti-oxidative enzymes Cu/Zn-SOD, Mn-SOD, and CAT were significantly inhibited by Ang II stimulation compared with controls and UDCA pretreatment up-regulated these enzymes after Ang II challenged, the effects were also through Nrf2 (Fig. 5D). The anti-oxidative effect of UDCA via Nrf2 was confirmed by detection of ROS production in the presence of Nrf2-siRNA by flow cytometry (Fig. 5E, 5F).

These results suggested that UDCA increased the expression of Nrf2 in VSMCs, thereby inhibiting Ang II-induced oxidative stress.

The apoptosis of VSMCs detected by FACS revealed that when the level of Ang II-induced ROS reduced by UDCA, the apoptosis level of VSMCs was also attenuated, the effect of which was comparable with NAC, indicating UDCA may affect the apoptosis of VSMCs associated with anti-oxidation (Fig. 6A).

The apoptotic index of VSMCs was also examined by expression of the apoptosis-associated proteins Bcl-2, Bcl-xl and Bax. The western blotting showed that Ang II resulted in lower Bcl-2 and Bcl-x expression and higher Bax expression compared with controls. These changes were reversed by UDCA treatments (Fig. 6B), the effects of which were in consistence with findings in ApoE^-/- mice (Fig. 6B). Another aspect, the mitochondrial membrane potential was detected by using JC-I and PI/FITC. The results of JC-1 and PI/FITC measured by FACS showed that UDCA significantly reduced Ang II-induced apoptosis of VSMCs (Fig. 6C).

In the presence of Nrf2-siRNA, the anti-apoptotic effect of UDCA after Ang II challenge examined by FACS was not observed, suggesting UDCA via Nrf2 involved in the anti-apoptotic effect (Fig. 7A). Western blotting also showed UDCA attenuated Ang II-induced VSMCs Apoptosis through Nrf2 signaling pathway. Ang II stimulation significantly decreased the expression of Bcl-2 and Bcl-xl, meanwhile increased Bax, the effect of which was reversed by pretreatment with UDCA. Adding Nrf2-siRNA abolished the anti-apoptotic effect of UDCA (Fig. 7B). The results of JC-1 and PI/FITC measured by FACS showed that UDCA significantly reduced Ang II-induced apoptosis of VSMCs through Nrf2 signaling pathway by using Nrf2-siRNA (Fig. 7C). These results suggested that UDCA attenuated Ang II-induced apoptosis of via VSMCs Nrf2.

The current study on UDCA showed important protective effects against Ang II-induced formation of AAD in ApoE^-/- mice. The occurrence of AAD and the maximal aortic diameter were significantly lower with treatment of UDCA. Oxidative stress plays an important role in the pathogenesis of many vascular diseases [19,20] including AAD [21]. Previous studies have indicated that UDCA, which is regarded as an antioxidant with documented efficacy in mice, is able to increase cell survival and decreases cellular apoptosis via anti-oxidative stress [24]. In our study, UDCA antagonized Ang II-induced oxidative stress via up-regulation of Nrf2, which involved in the regulation of NADPH subunits (p47, p67 and gp91) and redox enzyme (Cu/Zn-SOD, Mn-SOD and CAT), finally ameliorating apoptosis of VSMCs.

Our study provides some insight into the interpretation of whether oxidative stress contributes to formation of AAD in the Ang II-infusion model. A previous study showed that the hepatoprotective action of UDCA was partially dependent on its antioxidative effect [25]. UDCA has total antioxidative capabilities, including affecting the activitiy of SOD, catalase, MDA and total sulfhydryl content [26]. In addition, UDCA has been demonstrated that it was capable to affect another potent anti-oxidant Nrf2 in the development of oxidative stress [11,27]. As a basic-leucine zipper transcription factor, Nrf2 has been shown to regulate gene expression by binding to the antioxidant responsive element (ARE). The target genes of Nrf2 include NADPH quinine oxidoreductase-1 (NQO-1), hemeoxygenase-1 (HO-1) and glutathione S-transferase (GST) [28]. Previous study has proved that a positive association was existed between Nrf2 and redox enzymes (Cu/Zn-SOD, Mn-SOD and CAT) [29]. Thus, to investigate the role of UDCA in the production of ROS, we examined the expression of NADPH oxidase, catalase and SOD in aortic tissue from Ang II-induced AAD mice and Ang II-induced AAD mice treated with UDCA. In the experiments in vitro, Nrf2-siRNA was applied to detect whether the effect of UDCA was in Nrf2 dependent manner. As expected, pre-treatment of UDCA markedly ameliorated markers of oxidative stress and those redox enzymes in Nrf2 dependent manner, further inhibiting apoptosis of aortic medial cells which were induced by Ang II.

The mechanism that ROS levels are locally increased in AAD and contribute to the pathogenesis of this disease has been suggested from several observations. ROS can promote proliferation or apoptosis of VSMCs, and the role of oxidative stress depends on the relative degree of ROS [30]. In pathological conditions, high ROS level directly or through further stimulate inflammatory reaction-induced apoptosis [31]. Studies have shown that VSMCs apoptosis induced by oxidative stress increases the risk of dissection [32]. ROS have been reported to stimulate VSMCs to release a large amount of cyclophilinA, which enhances the local inflammatory response to aggravated oxidative stress. The present findings clearly indicate that ROS can effectively activate matrix metalloproteinases [33], which play a critical role in the pathogenesis of AAD, and substantially lead to degradation of the extra-cellular matrix.

In vascular cells, Ang II irritates the production of ROS via NADPH oxidase and has proapoptotic effects [34]. Our cellular observations further showed that UDCA suppressed Ang II-induced oxidative stress of VSMCs. The pivotal pathological feature in thoracic aortic dissection is ongoing medial degeneration of the aortic wall, which is characterized by VSMCs apoptosis [35]. Therefore, we investigated whether antiapoptotic action of UDCA is mediated by increased production of ROS in VSMCs. UDCA remarkably ameliorated Ang II-induced apoptosis of VSMCs, which was associated with a reduced level of oxidative stress. Besides VSMCs, other cells such as macrophages [15,36], mast cells [37], immune cells [37,38,39] also played very important roles in the development of pathological abdominal aortic aneurysms. Attenuation of the influx of inflammatory cells into the adventitia was proved to be associated with reduced mouse abdominal aortic aneurysm formations [40,41]. In our current studies, we found UDCA could play its role both on VSMCs and other cells infiltrated into the adventitia, through reducing ROS productions and apoptosis of targeted VSMCs and inflammatory cells, thereby attenuated the formation of AAA. Additionally, the inflammation in adventitial also influenced the function of VSMCs or neointima and the degradation of medial layer was the main cause of AAA rupture [42,43], thus, we mainly performed our experiments in VSMCs in vitro, although UDCA also affect the inflammatory response in adventitia.

Hypertension is an important risk factor in the genesis of AAD in humans [44]. Ang II also markedly elevates blood pressure in the AAD model. However, in our study, UDCA administration did not attenuate Ang II-induced hypertension. This result is consistent with the effects of vitamin E, a dietary antioxidant [45]. The lipid profiles of UDCA-treated mice were also unchanged in our study. Taken together, these results suggest that the protective effect of UDCA on Ang II-induced AAD is independent of lipid levels and lowering of blood pressure.

In summary, our study shows that UDCA protects against formation of AAD through the anti-oxidant Nrf2 in an Ang II-infused model in ApoE^-/- mice. These findings are consistent with the concept that there is a link between oxidative stress and pathogenesis of Ang II-driven development of AAD in ApoE^-/- mice. Furthermore, our results raise the possibility that preventive administration of UDCA and other antioxidants to patients at risk with AAD could be useful for reducing the occurrence of AAD.

The reported work was supported in part by research grants fromNational Nature Science Foundation of China (No. 81170259). Thanks for the assistance of Dr. Dao Wen Wang (The Institute of Hypertension and Department of Internal Medicine, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology) for the support of this work.

The authors declare that there is no conflict of interest.

W. Liu and B. Wang contributed equally to this work.