|Accueil | Aperçu | Revues | Soumettre | Nous Contacter | English|
We have previously reported that activation of AMP-activated kinase alpha 2 (AMPKa2) by nicotine or angiotensin II (AngII) instigates formation of abdominal aortic aneurysms (AAA) in Apoe−/− mice. Statins, used to treat hyperlipidemia widely, activate AMPK in vascular cells. We sought to examine the effects of pravastatin on AAA formation and uncover the molecular mechanism. The AAA model was induced by AngII and evaluated by incidence, elastin degradation, and maximal abdominal aortic diameter in Apoe−/− mice. The phosphorylated levels of AMPKa2 and activator protein 2 alpha (AP-2a) were examined in cultured vascular smooth muscle cells (VSMCs) or in mice. We observed that pravastatin (50 mg/kg/day, 8 weeks) remarkably increased the AngII-induced AAA incidence in mice. In VSMCs, pravastatin increased the levels of pAMPK, pAP-2a, and MMP2 in both basal and AngII-stressed conditions, which were abolished by tempol and compound C. Pravastatin-upregulated MMP2 was abrogated by AMPKa2 or AP-2a siRNA. Lentivirus-mediated gene silence of AMPKa2 or AP-2a abolished pravastatin-worsened AAA formations in AngII-infused Apoe−/− mice. Clinical investigations demonstrated that both AMPKa2 and AP-2a phosphorylations were increased in AAA patients or human subjects taking pravastatin. In conclusion, pravastatin promotes AAA formation through AMPKa2-dependent AP-2a activations.
Abdominal aortic aneurysms (AAAs) are permanent dilations of the abdominal aorta with over 85% mortality after rupture. There are no therapeutic strategies but surgery proven to blunt AAA progression and rapture. Pathologically, AAA is characterized by dilatation of all layers of the arterial wall as a result of loss of elastin, vascular smooth muscle cell (VSMC) apoptosis, and compensatory collagen deposition [1–3]. For example, angiotensin II (AngII) increases matrix metalloproteinases (MMPs) to induce AAA formation in animal experiments by upregulating oxidative stress [4, 5].
Statins are cholesterol-lowering drugs and widely used in the treatment of hypercholesterolemia to prevent the development of atherosclerosis through improving endothelial function [6–8]. However, statins produce several adverse effects, such as insulin resistance , skeletal muscle toxicity and myocardial atrophy . The AMP-activated protein kinase (AMPK), consisting of catalytic α subunit and regulatory subunits β and γ, has a pivotal function in energy homoeostasis in eukaryotes. AMPK is activated by pravastatin in skeletal muscle and endothelial cells [11, 12]. We reported that AMPK activation mediated AngII-induced AAA formation in Apoe−/− mice . This alludes that pravastatin is potentially to promote AAA incidences.
Activator protein 2α (AP-2α) is a member of the AP-2 transcription factor family consisting of α, β, γ, δ, and ε subunits . Mice deficient in AP-2 die after birth due to the abnormal development, suggesting an important role of AP-2 in mammals [15, 16]. Park et al reported that nerve growth factor stimulated endothelial cell invasion by augmenting MMP-2 via AP-2α-dependent gene transcription, which may be responsible for triggering angiogenesis . We found that aspirin activates AMPK to increase AP-2α in accelerated atherosclerotic plaque . This study is aimed to investigate the effects of pravastatin on AAA formations. Our results demonstrate that pravastatin augments AngII-induced AAA incidence through AMPKα2/AP-2α signaling.
As reported [19, 20], we found that treatment with AngII (1.44 mg/kg/day) for 28 consecutive days promotes AAA formation in Apoe−/− mice as evidenced by enlarged abdominal aortas morphologically (Figure (Figure1A).1A). To determine the effects of statin on AAA formation, we treated AngII-infused Apoe−/− mice with pravastatin (50 mg/kg/day) for 8 weeks. Compared with mice treated with vehicle, pravastatin significantly increased the AAA incidence (60.87% VS 91.30%, P<0.05), the mortality and the severity of AAA in AngII-infused mice (Figure (Figure1B1B and and1C).1C). Histological analysis also indicated that the maximal aortic diameter and the degradation of elastin were increased by pravastatin in Apoe−/− mice with AAA (Figure 1D-1F). These results suggest that pravastatin promotes the AngII-induced AAA formation in Apoe−/− mice.
We also detected whether pravastatin affected hemodynamic parameters or metabolic indexes in AngII-infused Apoe−/− mice. As indicated in Supplementary Table S1 and S2, pravastatin treatment had no effects on heart rates, blood pressures, plasma triglyceride and glucose levels in AngII-infused Apoe−/− mice. Interestingly, the levels of total cholesterol and LDL cholesterol, but not HDL cholesterol, were increased by pravastatin in Apoe−/− mice, demonstrating that the effect of pravastatin on AAA formation is not due to the lipid-lowering effects.
We previously reported that AMPK is activated by oxidants via threonine 172 phosphorylation (pAMPK-T172) and AMPK activation instigates AAA formation [13, 21]. Then we investigated whether pravastatin via oxidant activated AMPKα2. As indicated in Figure 2A-2C, the levels of pAMPK-T172 were gradually increased by pravastatin in concentration-dependent manner in cultured murine VSMCs. Tempol, as a SOD mimic, abolished the augments of Nox4, p47, 4-HNE, and pAMPK-T172 induced by pravastatin. Similarly, the ROS productions, assayed by determining the intensity of DHE fluorescence, were increased by pravastatin, which were abolished by tempol (Figure (Figure2D2D and and2E).2E). Taking all data together, it indicates that pravastatin via induction of oxidative stress activates AMPKα2 in VSMCs.
We have identified AP-2α as a downstream of AMPKα2 via phosphorylation at serine 219 (S219), which represents its transcriptional activity . Thus, we detected whether pravastatin increases AP-2α S219 phosphorylation. As depicted in Figure Figure3A,3A, pravastatin did not increase the level of pAP-2α-S219 at a concentration of 0.1-1 μM; however, at 10 μM, pravastatin significantly enhanced AP-2α S219 phosphorylation. Increasing concentrations of pravastatin (50-100 μM) further enhanced AP-2α phosphorylation. Further, reduction of ROS by tempol abolished pravastatin-induced AP-2α S219 phosphorylation (Figure (Figure3B),3B), revealing that pravastatin is a potential activator of AP-2α through ROS.
To determine the role of AMPKα2 in pravastatin-activated AP-2α, we pretreated VSMCs with compound C (20 μM) which functions as an AMPK inhibitor (Supplementary Figure S1A). As presented in Figure Figure3B,3B, pravastatin (50 μM) for 2 hours enhanced AP-2α phosphorylation in vehicle-treated cells, but not in VSMCs pretreated with compound C.
The effects of compound C on AP-2α phosphorylation were further confirmed by transfecting VSMCs with AMPKα2 siRNA to inhibit AMPKα2 expression (Supplementary Figure S1B). AMPKα2 siRNA, but not control siRNA, blocked the effects of pravastatin on AP-2α phosphorylation (Figure (Figure3C),3C), suggesting that the pravastatin-increased AP-2α phosphorylation is AMPKα2 dependent.
To validate how activation of AP-2α by pravastatin increases the AAA formation, we detected the levels of MMP2, which is a key contributor for elastin degradation in the process of AAA and a target of AP-2α in endothelial cells . In Figure Figure4A,4A, pravastatin concentration-dependently increased the protein levels of MMP2 in VSMCs. As expected, either tempol (Figure (Figure4B)4B) or compound C (Figure (Figure4C)4C) significantly abolished the pravastatin-enhanced the levels of MMP2 protein and activity, and mRNA (Supplementary Figure S1C).
To further confirm the essential roles of AMPKα2 and AP-2α in pravastatin-induced upregulations of MMP2, endogenous expressions of AMPKα2 and AP-2α in VSMCs were suppressed by transfection of AMPKα2 and AP-2α siRNA (Supplementary Figure S1B and S1D). Deficiency of AP-2α had no effects on pravastatin-increased AMPK phosphorylation, indicating that AP-2α is a downstream of AMPK (Supplementary Figure S1D). As shown in Figure Figure4D,4D, ,4E,4E, S1E, S1F, pravastatin increased the levels of MMP2 protein, mRNA and activity in cells transfected with control siRNA. All these effects produced by pravastatin were bypassed by knockdown of AMPKα2 and AP-2α, demonstrating that both AMPKα2 and AP-2α are required for pravastatin-induced MMP2 upregulation in VSMCs.
Knowing that pravastatin activated the AMPKα2/AP-2α/MMP2 signaling in VSMCs under resting condition, we next examined whether pravastatin activates the AMPKα2/AP-2α/MMP2 signaling in VSMCs with AngII. As indicated in Figure 5A-5C, both AngII and pravastatin alone increased the levels of pAMPK, pAP-2α, and MMP2 protein and mRNA expressions in VSMCs. Importantly, co-incubation of pravastatin further increased the levels of pAMPK, pAP-2α, and MMP2 protein and mRNA expressions in AngII-treated VSMCs, suggesting that pravastatin activates AMPKα2/AP-2α signaling in AngII-stressed cells.
To show the translational applicability of these observations from mice, we also tested whether they can be replicated in human VSMCs. Thus, primary human aortic VSMCs were incubated with AngII plus pravastatin. Similarly, pravastatin increased the levels of pAMPK, pAP-2α, and MMP2 protein and mRNA expressions in human VSMCs with or without AngII (Figure 5D-5F), implying that pravastatin activates AMPKα2/AP-2α/MMP2 signaling in humans.
The effects of pravastatin on oxidative stress, AMPKα2 and AP-2α phosphorylations, and MMP2 gene expression were further investigated in vivo. As indicated in Supplementary Figure S2A-S2C, in AngII-infused Apoe−/− mice, pravastatin administration dramatically increased the levels of 4-HNE, p47, Nox4, pAMPK-T172, pAP-2α-S219, and MMP2 protein and activity, compared to vehicle-treated mice, suggesting that pravastatin induces oxidative stress and activates AMPKα2/AP-2α/MMP2 signaling in vivo.
The role of AMPKα2 in pravastatin-increased AngII-induced AAA formation was also determined in Apoe−/− mice. Lentivirus-mediated RNA interference significantly inhibited AMPKα2 protein expression in abdominal aortic artery (Supplementary Figure S3B-S3E), but not affected hemodynamic parameters or metabolic indexes in AngII-infused Apoe−/− mice (Supplementary Table S3 and S4). Similarly, pravastatin significantly promoted the AngII-induced AAA formation in Apoe−/− mice infected with lentivirus expressing negative control shRNA, but not in Apoe−/− mice infected with lentivirus containing AMPKα2 shRNA (Figure 6A-6E). Collectively, it indicates that pravastatin via AMPKα2 activation promotes AAA formation in mice.
We next examined the roles of AP-2α in the effects of pravastatin on promoting AAA formation in Apoe−/− mice. Due to the lethal phenotype of mice deficient in AP-2α , we generated AP-2α-knockdown mice by infecting lentivirus containing AP-2α short hairpin RNA (shRNA) to downregulate AP-2α protein expression in Apoe−/− mice (Supplementary Figure S3B-S3E), which did not affect hemodynamic parameters or metabolic indexes in AngII-infused Apoe−/− mice (Supplementary Table S5 and S6). As depicted in Figure Figure7A,7A, pravastatin treatment significantly accelerated the AngII-induced AAA formation in Apoe−/− mice infected with lentivirus expressing scramble shRNA. The incidence and mortality of AAA (Figure (Figure7B),7B), maximal diameter (Figure (Figure7C),7C), and elastin degradation (Figure (Figure7D7D and and7E)7E) in abdominal aortic artery were increased, compared to vehicle-treated mice. However, all these effects induced by pravastatin were limited in Apoe−/− mice if they were infected with lentivirus expressing AP-2α shRNA. Together, these data demonstrate that AP-2α is required for pravastatin-enhanced AAA formation in mice.
The pathophysiological feather of AAA is the elastin degradation in vascular medium, which is mediated by MMPs (MMP2 and MMP9) . We then examined the effects of AMPKα2 or AP-2α shRNA on MMP2/9 in vivo. As shown in Supplementary Figure S4A-S4B and S5A-S5D, the levels of MMP2 mRNA, protein, and activity were remarkably increased in Apoe−/− mice expressing scramble shRNA by pravastatin administration, compared to vehicle treated mice. However, either AP-2α or AMPKα2 shRNA inhibited the enchantment of MMP2 induced by pravastatin in AngII-infused Apoe−/− mice. The levels of MMP9 protein in cross-sections of the abdominal aortas were comparable among each group, indicating that MMP2 plays a dominant role in pravastatin-induced elastin degradation, similar to the report from Bradford C Berk's lab . Taken these data together, it reveals that AMPKα2 or AP-2α is crucial for pravastatin-enhanced MMP2 activation in vivo.
To address if pravastatin at conventional dose (20 mg/day) activates AMPKα2/AP-2α signaling, we assayed the phosphorylated levels of AP-2α and AMPKα2 in peripheral blood leucocytes from human individuals taking pravastatin (Supplementary Table S7). In Figure Figure8A,8A, the levels of pAP-2α-S219 and pAMPK-T172 were remarkably increased by taking pravastatin in humans, accompanied with the enhancement of serum MMP2 levels (Figure (Figure8B).8B). These data indicate that pravastatin is able to activate AMPKα2/AP-2α/MMP2 signaling in humans at regular dose.
Finally, to establish human relevance, we conducted a pilot experiment by collecting samples of AAA from human subjects (Supplementary Table S8). As shown in Figure 8C-8E, human AAA samples exhibited higher levels of AMPKα2 and AP-2α phosphorylations in aortic tissues from AAA than those from non-AAA control patients. Although the pilot experiment did not establish the cause-effect relation among AP-2α and AAA in clinical investigations, it still implies the important role of AP-2α in the process of AAA.
The major finding of this study is that pravastatin promotes the AAA formation. To date, the effects of statins on AAA remain controversial. Simvastatin has been reported to inhibit AAA formation induced by angiotensin II (AngII) in apolipoprotein E knockout (Apoe−/−) mice [24–26], while rosuvastatin and atorvastatin had no benefit for abdominal aortic aneurysms . Pravastatin produces bidirectional effects on cerebral aneurysm in estrogen-deficient rats, which is associated with the dose . Karrowni et al retrospectively found that statin therapy led to decreased AAA growth rate in humans, and while meta-analyses and individual studies have failed to show benefit or shown mildly reduced growth . Hurks et al reported that there was a trend towards increased aortic MMPs and protease activity in AAA patients taking pravastatin, but not atorvastatin or simvastatin . The distinct effects of statins on AAA may depend on the subtypes, doses, durations, and AAA models.
We also identified a novel molecular mechanism that AP-2α S219 phosphorylation is crucial for pravastatin-promoted AAA formation. As a transcription factor, AP-2α recognizes the consensus DNA sequence of 5′-GCCNNNGGC-3′, which regulates a number of genes including MMP2, p21, Bcl-2, and VEGF, which are involved in multiple cell functions. We previously identified AMPKα2 as an upstream kinase of AP-2α . In this study, we further discovered that statins function as AP-2α activator through AMPKα2.
The results presented in this work have for the first time established a causal effect of statins on AAA in vivo and uncovered the molecular mechanism by which pravastatin promotes AAA formation (Supplementary Figure S6). We suggest that pravastatin administration would be cautiously reconsidered in patients with high serum level of AngII or low blood level of apolipoprotein E, and patients with AAA.
Detailed information is available in Supplementary Materials.
Male Apoe−/− mice were received pravastatin administration (50 mg/kg) in drinking water. The model of AAA was established by implanting an Alzet osmatic minipumps into Apoe−/− mice to deliver AngII (1.44 mg/kg/day, 28 days) as described previously . Mice were injected with GFP-labeled lentivirus expressing AP-2α or AMPKα2 shRNA via tail vein as described previously . At the end of experiment, all mice were sacrificed under anesthesia. The whole aortas including thoracic and abdominal aortas were collected for morphological and histological analysis of AAA.
To quantify AAA incidence and size, the maximum width of the abdominal aorta was measured with Image Pro Plus software (Media Cybernetics Inc.). Aneurysm incidence was defined as an external width of the suprarenal aorta that was increased by 50% or greater as described previously .
AAA tissues were obtained from human subjects under surgery. Patients were enrolled to take pravastatin (20 mg/kg/day) for 0.5-3 years. The blood was collected from human subjects before and after pravastatin treatment. Leucocytes were isolated from blood as described previously .
Based on the protocol from Signaling Gateway, the shRNA cassette containing target sequence of AP-2α (GGAGAGCGAAGTCTAAGAATG) or AMPKα2 (GGAGAGCUAUUUGAUUAUATT) was designed. The cassette was subcloned into pEN-hH1c vector as described previously .
As described previously , mouse and human vascular smooth muscle cells (VSMCs) from ATCC were grown in basal medium (Clonetics Inc. Walkersville, MD) supplemented with 2% fetal bovine serum, penicillin (100 U/ml) and streptomycin (10 mg/ml). In all experiments, cells were used between passages 4 and 8. All cells were incubated at 37°C in a humidified atmosphere of 5% CO2 and 95% air. Cells were grown to 80% confluency before being treated with different agents.
Transient transfection of siRNA was carried out according to Santa Cruz's protocol .
Zymography was completed by use of a MMP gelatin zymography kit (GenMed Scientific Inc., USA) as described previously .
For the relative quantitation of western blot, qPCR and zymography, the intensity of bands were calculated and the background was subtracted from the calculated area. We setup the ration of control group as 1. All quantitative data were reported as mean ± SEM. One-way ANOVA followed by Tukey post-hoc tests was used for multiple comparisons. Chi-Square test was used for statistical comparisons of AAA incidence, mortality, and severity. Two-sided P-value <0.05 was considered as significant.
CONFLICTS OF INTEREST
There is no conflict of interest.
This work was supported by National 973 Basic Research Program of China (2013CB530700), National Natural Science Foundation of China (81320108004, 81370411, 81470591, 81570723, 81570729, 81673423), Natural Science Foundation of Shandong Province (ZR2016HM50), and Natural Science Foundation of Henan Province (162300410216, 121100910300, 132102310247). This project was also sponsored by Program for New Century Excellent Talents in University (NCET-13-0351), Program of Clinical Investigation (Nanshan Group), Qilu Hospital, Shandong University (2014QLKY15), and the Research Fund of Xinxiang Medical University (2016PN-KFKT-02, XYBSKYZZ201626, 2014QN153, ZD2011-30). Shuang-Xi Wang is a recipient of Qilu Professional Scholar of Shandong University and a Taihang Professional Scholar of Xinxiang Medical University (505067). Dr. Ya-Ling Yin is a recipient of Taihang Young Scholars of Xinxiang Medical University.