α1AMP activated protein kinase mediates the vascular protective effects of exercise

Abstract

Objective

We investigated whether adenosine monophosphate activated protein kinase (AMPK) may be involved in the signaling processes leading to exercise-mediated vascular protection.

Methods and Results

The effects of voluntary exercise on AMPK activity, endothelial nitric oxide synthase (eNOS) expression and phosphorylation, vascular reactive oxygen species (ROS) formation and cell senescence were tested in α1AMPK knockout and corresponding wildtype mice. Exercise significantly improved endothelial function and increased plasma nitrite production in wildtype mice, associated with an activation of aortic AMPK assessed by its phosphorylation at threonine172. In addition, regular physical activity resulted in an upregulation of eNOS protein, serine1177 eNOS phosphorylation and an increase of circulating Tie-2⁺Sca-1⁺Flk-1⁺ myeloid progenitor cells. All these changes were absent after α1AMPK deletion. In addition, exercise increased the expression of important regulators of the antioxidative defense including heme oxygenase 1 (HO-1) and peroxisome proliferator-activated receptor gamma coactivator-1α (PGC-1α), decreased aortic ROS levels and prevented endothelial cell senescence in an α1AMPK-dependent manner.

Conclusions

Intact α1AMPK signaling is required for the signaling events leading to the manifestation of vascular protective effects during exercise. Pharmacological AMPK activation might be a novel approach in the near future to simulate the beneficial vascular effects of physical activity.

Introduction

AMP-activated protein kinase (AMPK) is considered to be a metabolic master switch, which governs energy expenditure and energy production according to demand. It is activated by decreased ATP levels and a concomitant rise in cellular AMP, which allows cellular survival during metabolic stress. Emerging data suggests that AMPK has also distinct functions in the vasculature as it activates and phosphorylates endothelial nitric oxide synthase¹, protects endothelial cells against oxidative stress², prevents vascular smooth muscle proliferation³ and mediates angiogenesis⁴. All these effects suggest a protective role of AMPK in the vascular system.

Sedentary lifestyle is a major cause for the increase of vascular diseases in Western societies, and exercise is a simple and effective way to prevent cardiovascular risk factors such as obesity, diabetes and hypertension in the long term⁵. Upon physical activity, AMPK is activated in skeletal muscle and recent data suggest that its activation is required for the metabolic response to exercise in vivo⁶. Besides the modulation of cardiovascular risk factors, exercise may also have direct effects on the vasculature leading to an improvement of endothelial function⁷. The mechanisms how exercise may directly affect vascular function are rather multifactorial and include an upregulation of endothelial nitric oxide synthase (eNOS) expression⁸, serine1177 eNOS phosphorylation⁹, decreased oxidative stress e.g. by inhibition of vascular NADPH oxidases¹⁰ or augmented superoxide dismutase (SOD) expression¹¹, increased number of endothelial progenitor cells and enhanced angiogenesis¹². Increased cardiac output and blood flow during exercise will enhance forces that act on the endothelium¹³, and shear stress in particular is known to be one of the strongest stimuli for eNOS activation¹⁴. However, so far no unifying signaling event has been characterized that is able to initiate all these beneficial events in response to exercise. Since endothelial AMPK is activated in response to shear stress¹⁵ and exhibits similar vascular protective effects as compared to exercise, we investigated whether exercise activates AMPK in the vasculature and whether AMPK may be involved in the signaling processes leading to exercise-mediated vascular protection.

Materials and Methods

Animals and exercise protocol

All animal treatment was in accordance with the Guide for the Care and Use of Laboratory Animals as adopted and promulgated by the U.S. National Institutes of Health and was granted by the Ethics Committee of the University Hospital Mainz. To study the role of AMPK for the vascular effects of exercise, α1AMPK knockout mice and corresponding littermate wild type mice (C57Bl6/129Sv/FVB-N background) were used¹⁶. Eight-week-old male mice were kept in individual cages for eight weeks equipped with a running wheel and a mileage counter. Body weight and heart weight were assessed in all animals.

Reagents

Antibodies against ser79p-ACC, total-AMPK and thr172p-AMPK were from Cell Signaling (Boston, MA, USA). The total-eNOS, ser1177p-eNOS, Chk-2, p53, p16^INK4, Sca-1-FITC and Flk-1-PE antibodies were purchased from BD biosciences (San Jose, CA, USA). The Tie-2 PE antibody was purchased from ebiosciences (San Diego, USA). All other chemicals and reagents were of analytical grade purchased from Sigma-Aldrich, Fluka, or Merck.

Isometric Tension Studies

Vasodilator responses to acetylcholine (ACh) were assessed with endothelium-intact isolated mouse aortic rings (thoracic aorta, 3mm in length) mounted for isometric tension recordings in organ chambers, as described previously¹⁷.

Determination of NO synthesis via plasma nitrite

Nitrite, the oxidation product of NO, was assayed in mouse plasma as an indicator of NO synthesis. Nitrite was determined after chemical reduction to NO and subsequent reaction with ozone by chemiluminescence using a NOATM 280 Nitric Oxide Analyzer (Sievers)^{18, 19}. Nitrite plasma levels correlate well with NO biosynthesis^{20, 21}.

Fluorescence-Activated Cell Sorter Analysis

Peripheral mouse blood was analyzed as described previously^{12, 22}. The viable lymphocyte population was gated for Tie-2-PE positive cells, which were analyzed for Sca-1-FITC and Flk-2-APC double positive cells²². Tie-2⁺Sca-1⁺Flk-1⁺ cells were defined as EPC’s.

Isolation of bone marrow derived EPCs and Colony forming units (CFU) assay

Isolation of bone marrow derived EPCs was performed by crushing the compact bones in a mortar containing PBS with 2% FBS. Cells were filtered through a 70 μm cell strainer and incubated with collagenase type 1 purchased from Sigma Aldrich. Colony forming units were assessed as described previously^{23, 24}. Briefly, bone marrow derived cells were cultured in endothelial cell specific medium for 3 days, washed and transferred in ColonyGEL^™ 1202 Methylcellulose (CellSystems, Troisdorf, Germany) containing Mouse VEGF (Milteny Biotec, Bergisch Gladbach, Germany). Colonies were counted after two weeks of culture.

Serum antioxidant capacity

Acetonitrile was added to freshly prepared serum in equal volumes, incubated for 4 minutes at room temperature and centrifuged at 9500 × g at 4°C. The assay was started by addition of the deproteinized serum to a 50μM 2,2-diphenyl-1-picryl-hydrazyl radical (DPPH) solution and was measured at 517 nm during a time period of 30 minutes. Antioxidants will reduce the DPPH radical resulting in the loss of its absorbance at 517 nm. Serum antioxidant capacity was calculated by the decrease in the absorbance of the DPPH radical as described previously²⁵.

Vascular reactive oxygen species (ROS) production

The fluorescent dye dihydroethidine (DHE) was used to detect vascular superoxide production in situ as described previously²⁶. In intact aortic rings, ROS levels including superoxide, hydrogen peroxide and peroxynitrite were analyzed by L-012 enhanced chemiluminescence.

Reverse transcription real-time PCR (qRT-PCR)

mRNA expression was analyzed by quantitative real-time RT-PCR as previously described¹⁷. Briefly, total RNA from mouse aorta was isolated (RNeasy Fibrous Tissue Mini Kit; Qiagen, Hilden, Germany), and 50ng of total RNA was used for real-time RT-PCR analysis with the QuantiTect^™ Probe RT-PCR kit (Qiagen). A TaqMan® Gene Expression assay for TBP, TERT, TRF-2, Nrf-2, PGC-1α, HO-1 was purchased as probe-and-primer set (Applied Biosystems, Foster City, CA). The comparative ΔΔCt method was used for relative mRNA quantification²⁷. Gene expression was normalized to the endogenous control, TATA box binding protein (TBP) mRNA, and the amount of target gene mRNA expression in each sample was expressed relative to that of wildtype.

Immunoblotting

Isolated aortic tissue was frozen and homogenized in liquid nitrogen. Proteins were separated by SDS-Page, blotted onto nitrocellulose membranes and immunoblotting was performed as described previously².

Statistical Analysis

Results are expressed as mean ± S.E.M. One-way analysis of variance (with Bonferroni or Dunn correction for comparison of multiple means) was used for comparisons of vasodilator potency and efficacy and vascular superoxide production. The EC₅₀ value for vascular reactivity studies was obtained by log-transformation. p-values < 0.05 were considered significant.

Results

Running distance and biometrical data

The mean running distance in wildtype mice was 4336 ± 842 m per 24 hours and did not differ significantly in α1AMPK knockout mice (“A”, 4002 ± 1200 m). Body weight showed a significant decline in the exercise group, which was comparable in wildtype and α1AMPK knockout mice (Fig. 1B–C).

Figure 1. Effects of exercise on body weight, heart weight and vascular AMPK activity.

Eight-week-old male mice were kept in individual cages for eight weeks equipped with a running wheel and a mileage counter. Running distance is displayed as mean ± SEM of n= 15 – 20 (A). Body weight (B) and heart weight/body weight ratio (C) were assessed in all animal groups at the end of the exercise period; data are mean ± SEM of n= 15 – 20. Aortic tissues were homogenized and the lysates corrected for protein content. Immunoblotting was performed using antibodies against serine79p-ACC (E), threonin172p-AMPK (F), ACC (G), total αAMPK (H) and. Alpha-actinin served as a loading control. The shown immunoblot is representative of 8 independent experiments (D). * indicates p<0.05 vs. WT, ** indicates p<0.05 vs. α1AMPK −/−, # indicates p<0.05 vs. WT + exercise.

Vascular AMPK activity is increased during chronic exercise

Exercise resulted in a significant increase in vascular AMPK activity as monitored by AMPK phosphorylation at threonine172 and phosphorylation of its downstream target acetyl CoA carboxylase (ACC) at serine79 (Fig. 1E–G). Deletion of α1AMPK lead to a 80% reduction in total AMPK expression (Fig. 1H) and a complete loss of vascular AMPK phosphorylation, confirming the prominent expression of the α1 containing AMPK in vascular cells.

Exercise improves endothelial function in an α1AMPK-dependent manner

Exercise lead to a marked improvement of endothelium dependent relaxation, while it had no effect on endothelial function in α1AMPK knockout mice (Fig. 2A). The degree of preconstriction after prostaglandin F_2α was comparable among all groups (Suppl. Fig. 2A). Since AMPK and exercise are both known to activate eNOS^{1, 8, 11}, we next investigated whether the improvement of endothelial function in wildtype mice was associated with increased vascular NO formation. As expected, plasma nitrite levels used as a surrogate for vascular NO production were increased during exercise in wildtype animals. In contrast, α1AMPK knockout animals had similar basal levels but showed no increase in response to exercise (Fig. 2B). In accordance with previous studies, improved NO signaling in wildtype animals was associated with increased eNOS protein expression and its phosphorylation at serine1177 (Fig. 2C–F). Because eNOS gene expression is mediated by nuclear factor erythroid 2-like 2 (Nrf2) in response to shear stress²⁸, this mechanism may also play an important role during exercise as hemodynamic changes will also increase the forces that act on the vascular endothelium. In fact we found increased aortic Nrf2 expression in response to exercise, an effect that was absent in the vasculature of α1AMPK knockout mice, suggesting that AMPK signalling must be upstream of Nrf2 (Fig. 2G). Although activation of Foxo3a was also shown to regulate eNOS expression in vivo²⁹ and AMPK may directly phosphorylate Foxo3a at Ser-413³⁰, we observed no significant change in Foxo3a phosphorylation among all treatment groups (Suppl. Fig. 1A). Taken together, our results strongly indicate that the blunted eNOS upregulation in α1AMPK knockout mice undergoing exercise is mediated by an attenuated Nrf2 expression, while Foxo3a did not modulate eNOS expression in our model. The decreased eNOS phosphorylation may be a direct consequence of diminished vascular AMPK activity, as eNOS is a known direct target of AMPK³¹.

Figure 2. Endothelial function and vascular NO production is enhanced by exercise in wildtype mice but not in α1AMPK knock out mice.

Isometric tension studies in intact aortic rings (3mm in length) were performed to assess endothelial function in response to acetylcholine (ACh) ex vivo (A). The average preconstriction obtained by PGF_2α 1μM was ~0,55g and similar between all groups. Data are means ± SEM of n=10 independent experiments. Vascular NO production was determined by serum nitrite levels using an NO analyzer (B), data are means ± SEM of n=10–14. Immunoblotting was performed in aortic homogenates using serine1177-phospho-eNOS (D) and total eNOS (E) antibodies. eNOS expression was normalized to PECAM-1, while eNOS phosphorylation was calculated by the ratio of serine1177-phospho-eNOS: total eNOS (F). The Immunoblot shown is representative of 8 independent experiments (C), bar graphs were obtained after densitometric analysis. Aortic mRNA Expression of Nrf2 was determined by reverse transcription real-time PCR (G), data are mean ± SEM of n=8 independent experiments. * indicates p<0.05 vs. WT, # indicates p<0.05 vs. WT + exercise.

Effects of α1AMPK deletion on antioxidant capacity and vascular oxidative stress during exercise

Endothelial function largely depends on the critical balance between vascular NO production and its inactivation by ROS, in particular superoxide anions. Since exercise is known to decrease ROS levels by an upregulation of antioxidative enzyme systems such as heme oxygenase 1 (HO-1)³², we next assessed serum antioxidant capacity by DPPH assay²⁵. Exercise resulted in a significant improvement of antioxidant capacity in wildtype mice, while it failed to increase the antioxidant defense in α1AMPK knockout mice (Fig. 3A). This was paralleled by an increased expression of PGC-1α (Fig. 3B), an important regulator of the cellular antioxidative defense³³, and an upregulation of HO-1 (Fig. 3C), which is known to be regulated in a Nrf2-dependent manner in endothelial cells³⁴. Since antioxidant serum capacity is a surrogate of systemic oxidative stress, we next examined vascular ROS levels. Our results show that exercise was able to reduce aortic ROS in wildtype mice but not in α1AMPK knockout mice as measured by L-012 enhanced chemiluminescence (Fig. 3D) and dihydroethidine staining (Fig. 3E). Due to the fast scavenging of superoxide anions by NO, increased NO production may contribute to the observed decline in vascular ROS levels. In order to evaluate the relative importance of reduced ROS levels for the improvement of vascular function by exercise, we performed additional isometric tension studies in aortic rings with addition of the superoxide scavenger polyethylene glycol-conjugated superoxide dismutase (PEG-SOD) in vitro. In wildtype mice, PEG-SOD improved endothelial function only in aortic rings from untreated but not from running mice, suggesting that exercise already decreased ROS levels and therefore no further improvement after PEG-SOD was observed. Aortic rings from AMPK knockout mice showed an improvement of endothelial function by PEG-SOD preincubation in both the control and exercise group (Suppl. Fig. 2B), compatible with increased basal oxidative stress in these animals. Endothelial dependent relaxation was completely blunted after eNOS inhibition with L-NAME in all groups (Suppl. Fig. 2C).

Figure 3. Exercise decreases systemic and vascular oxidative stress in an α1AMPK dependent manner.

Serum antioxidant capacity was measured by DPPH Assay (A). Data are mean ± SEM of n= 9 – 13. Aortic mRNA expression of PGC-1α (B) and HO-1 (C) was determined by reverse transcription real-time PCR, data are mean ± SEM of n=8. Aortic ROS production was assessed by L-012 enhanced chemiluminescence; data are mean ± SEM of n= 6 (D). Transverse aortic cryosections were labeled with dihydroethidine (DHE; 1 μmol/L), which produces red fluorescence when oxidized to 2-hydroxyethidium by superoxide. “E” indicates the endothelium (points to the right), “A” indicates the adventitia, lamina autofluorescence is green. (E). A summary of the densitometric analysis is shown on the right, data are representative of n=4. * indicates p<0.05 vs. WT, # indicates p<0.05 vs. WT + exercise.

Since the modulation of antioxidant enzyme systems such as PGC-1α and HO-1 goes in parallel with increased NO production (which may itself scavenge significant amounts of ROS), it seems that both of these α1AMPK-mediated effects contribute to a reduction of vascular ROS levels and finally an improvement of endothelial function during exercise.

Effects of α1AMPK deletion on vascular cell senescence and regeneration during exercise

Besides eNOS expression and activity, the balance between endothelial cell death and survival constitutes a distinctive factor of vascular function. In this respect, telomerase reverse transcriptase (TERT), the catalytic protein subunit of the enzyme telomerase and telomere repeat-binding factors (TRFs) are important components of the telomere complex, which is a central regulator of cell senescence³⁵. In accordance with a previous study³⁶, physical exercise resulted in an increased expression of the telomere-regulating proteins TRF2 and TERT (Fig. 4A&B) in the vasculature of wildtype animals, which was absent in α1AMPK knockout mice. In accordance with this observation, voluntary running wildtype mice showed a decreased protein expression of p53, cell-cycle-checkpoint kinase 2 (Chk2) and p16^INK4 in comparison to sedentary mice, compatible with decreased vascular aging (Fig. 4C–E). Deletion of α1AMPK resulted in a significant attenuation of these changes, indicating that the effects of exercise on cellular aging depend at least in part on the presence of α1AMPK. Immunohistochemistry revealed that exercise lead to a downregulation of p16^INK4 and Chk2 in particular in the endothelium of wildtype but not α1AMPK knockout mice, suggesting that the prevention of endothelial cell senescence by α1AMPK may considerably contribute to the increased NO bioavailability in wildtype mice undergoing exercise (Suppl. Fig. 3A&B).

Figure 4. Effects of exercise on vascular cell senescence.

Aortic mRNA Expression of TRF2 (A) and TERT (B) was determined by reverse transcription real-time PCR. For vascular p53 (C) Chk2 (D) and p16^INK4 (E) protein expression, a representative Immunoblot of n=6 independent experiments is shown; bar graphs were obtained by densitometry. Data are means ± SEM of n=6, * indicates p<0.05 vs. untreated WT; # indicates p<0.05 vs. WT + exercise; ** indicates p<0.05 vs. AMPK −/−. EPC were defined as Tie-2, Sca-1 and Flk-1 positive cells and their number was assessed by flow cytometry (F) and by CFU assay from bone marrow derived cells (G). Data are mean ± SEM of n=10–14. * indicates p<0.05 vs. WT, # indicates p<0.05 vs. WT + exercise.

Since putative bone marrow derived endothelial progenitor cells (EPCs) may replace injured endothelium, we also investigated the number of circulating Tie-2⁺Sca-1⁺Flk-1⁺ cells in α1AMPK knockout and corresponding wildtype mice. As described previously¹², the number of EPCs in wildtype mice was significantly increased during exercise while these cells contained comparable amounts of the α1AMPK vs. α2AMPK isoform (Suppl. Fig. 5). Compared to wildtype mice, α1AMPK knockout mice showed no significant increase of Tie-2⁺Sca-1⁺Flk-1⁺ cells in response to exercise (Fig. 4F). Similarly, the colony-forming capacity of EPCs - assessed by colony forming units (CFU) assay - was markedly increased in wildtype mice during exercise, while α1AMPK knockout mice showed no exercise-dependent increase (Fig. 4G, Suppl. Fig. 4). These results stress that AMPK is an important regulator of cell senescence and endothelial repair during chronic exercise training.

Discussion

In the current study, we provide evidence that AMPK activation is a key signaling event that mediates to a large part the protective effects of exercise in the vasculature. Deletion of the predominant vascular AMPK isoform α1AMPK prevented the manifestation of several protective effects of exercise including increased eNOS expression/ser1177-phosphorylation, decreased oxidative stress, an attenuation of cell senescence, and an increase of circulating Tie-2⁺Sca-1⁺Flk-1⁺ endothelial progenitor cells.

AMPK is usually known to be activated by stimuli that lead to increased ATP consumption and a concomitant rise in cellular AMP. In this respect, it is conceivable that AMPK is activated in skeletal muscle during exercise, since muscle work largely depends on rapid available energy stores in the form of ATP and creatin phosphate. Since AMPK can also improve cellular glucose uptake, it was postulated that the positive effects of exercise on glycemic control in diabetics are AMPK dependent³⁷. A recent report confirmed this hypothesis and showed that intact AMPK signaling is required for the metabolic response to exercise in vivo⁶. In the present study, we can extend these observations by showing that AMPK activation in response to exercise occurs also in the vasculature. It is tempting to speculate that an increasing AMP/ATP ratio is the initiating event to stimulate exercise-mediated AMPK activation in the vessel wall similar to the findings in skeletal muscle. Despite this appealing hypothesis, it remains to be established whether energy-dependent mechanisms are the primary means of AMPK regulation in the vasculature, in particular since upstream activators of AMPK such as LKB-1 have been described³⁸. Shear stress may be another important factor to drive AMPK activation in the vasculature, while NO itself may help to maintain AMPK activation in a positive feedback loop³⁹. Independent of its mode of activation, AMPK appears to be a very proximal signal in the adaptation to exercise, since our results demonstrate that AMPK is clearly upstream of Nrf2- and eNOS signaling and therefore establish AMPK as an early signaling event that mediates the protective effects of exercise in the vasculature (see proposed scheme in Fig. 5).

Figure 5. Proposed scheme to illustrate the AMPK-dependent vascular effects during exercise.

Chronic exercise will temporarily increase cardiac output and thereby augment shear stress that acts on the endothelial cell layer. Shear stress in turn activates AMPK, which mediates eNOS protein upregulation and its phosphorylation at serine1177. These events are paralleled by increased Nrf2 expression, which will lead to HO-1 upregulation. Together with an increased NO production, these events will lead to a reduction of vascular ROS levels. The effects of exercise on endothelial progenitor cell (EPC) number and endothelial function are AMPK-dependent, but also require intact eNOS signaling¹².

Physical exercise increases cardiac output and vascular perfusion, which will automatically increase shear stress that acts on the vascular wall. As a consequence, shear stress leads to eNOS activation and vasodilation, and it is attractive to speculate that endothelium-dependent processes are crucial for the manifestation of protective vascular effects in response to exercise. Our data implicate that AMPK and maybe Nrf2, which both are activated in response to shear stress⁴⁰, are part of a signaling cascade that leads to eNOS upregulation and vascular protection during exercise.

Previous studies regarding the effects of exercise on oxidative stress have yielded equivocal results. High endurance training was even shown to increase systemic oxidative stress, while moderate physical activity had no such effect⁴¹. These findings might also relate to the importance of antioxidant defense strategies in this setting, as exercise is able to increase HO-1³² or SOD activity¹¹. Our observations are in line with these findings since serum antioxidant capacity and the expression of HO-1, Nrf2 and PGC-1α were increased by exercise in an α1AMPK-dependent manner, leading to a significant decline in aortic ROS levels. In addition, the experimental setting with voluntary exercise in mice may have prevented pro-oxidative effects associated with forced exercise or high endurance training.

Many of the vascular effects in response to exercise have been described in a similar fashion for AMPK activation - including eNOS activation, protection against oxidative stress and angiogenesis - which prompted us to investigate the role of AMPK for the protective effects of exercise. In addition to the modulation of eNOS activity, endothelial function may also depend on the critical balance between endothelial cell senescence and regeneration, in particular by putative vascular progenitor cells. A previous study by Werner and coworkers demonstrated that exercise is able to prevent vascular cell senescence in a TERT- and eNOS-dependent manner³⁶. Here we demonstrate that AMPK is required to prevent cellular senescence by exercise, as AMPK deletion prevented exercise-induced increases of telomere regulating proteins and downregulation of Chk2, p53 and p16^INK4. Regarding endothelial repair, a potential mechanism is the replacement of injured endothelial cells by bone marrow derived endothelial progenitor cells⁴². A previous study demonstrated that AMPK plays a pivotal role in the differentiation of putative vascular progenitor cells⁴³. Here we demonstrate for the first time that α1AMPK regulates the number of circulating Tie-2⁺Sca-1⁺Flk-1⁺ myeloid progenitor cells in vivo. Mechanistically, α1AMPK may prevent oxidative damage of these cells in a similar fashion as previously described for endothelial cells², even though vascular progenitor cells have been reported to exhibit a higher resistance to oxidative stress⁴⁴. Since our study uses a global (and not tissue specific) knockout model, we can not rule out that α1AMPK deletion in other cell types besides endothelial cells or circulating EPC’s contribute to the observed vascular phenotype.

A landmark study by Narkar et al. demonstrated that AMPK activation has endurance enhancing properties⁴⁵, making it an attractive target for pharmacological interventions in order to simulate the beneficial effects of exercise. Current available strategies to activate AMPK pharmacologically include biguanides and glitazones. However, their AMPK activating properties are secondary to the inhibition of the mitochondrial respiratory chain⁴⁶ and therefore, the development of more potent and specific compounds remains a future goal. Our current results support the concept that the vascular protection through exercise is AMPK-dependent and that AMPK may be an attractive future pharmacological target to simulate the protective effects of exercise.

List of Abbreviations

ACC

acetyl CoA carboxylase

ACh

acetylcholine

AMP

adenosine monophosphate

AMPK

AMP activated protein kinase

Chk2

cell-cycle-checkpoint kinase 2

DHE

dihydroethidine

DPPH

2,2-diphenyl-1-picryl-hydrazyl radical

eNOS

endothelial nitric oxide synthase

EPC

endothelial progenitor cell

Foxo

Forkhead box O

HO-1

heme oxygenase 1

NO

nitric oxide

Nrf2

nuclear factor erythroid 2-like 2

PEG-SOD

polyethylene glycol-conjugated superoxide dismutase

PGC-1α

peroxisome proliferator-activated receptor gamma coactivator 1α

PGF_2α

prostaglandin F_2α

ROS

reactive oxygen species

SOD

superoxide dismutase

TBP

TATA box binding protein

TERT

telomerase reverse transcriptase

TRF-2

telomere repeat binding factor 2

VEGFR

vascular endothelial growth factor receptor