Acute Simvastatin Inhibits KATP Channels of Porcine Coronary Artery Myocytes

Background Statins (3-hydroxy-3-methyl-glutaryl coenzyme A (HMG-CoA) reductase inhibitors) consumption provides beneficial effects on cardiovascular systems. However, effects of statins on vascular KATP channel gatings are unknown. Methods Pig left anterior descending coronary artery and human left internal mammary artery were isolated and endothelium-denuded for tension measurements and Western immunoblots. Enzymatically-dissociated/cultured arterial myocytes were used for patch-clamp electrophysiological studies and for [Ca2+]i, [ATP]i and [glucose]o uptake measurements. Results The cromakalim (10 nM to 10 µM)- and pinacidil (10 nM to 10 µM)-induced concentration-dependent relaxation of porcine coronary artery was inhibited by simvastatin (3 and 10 µM). Simvastatin (1, 3 and 10 µM) suppressed (in okadaic acid (10 nM)-sensitive manner) cromakalim (10 µM)- and pinacidil (10 µM)-mediated opening of whole-cell KATP channels of arterial myocytes. Simvastatin (10 µM) and AICAR (1 mM) elicited a time-dependent, compound C (1 µM)-sensitive [3H]-2-deoxy-glucose uptake and an increase in [ATP]i levels. A time (2–30 min)- and concentration (0.1–10 µM)-dependent increase by simvastatin of p-AMPKα-Thr172 and p-PP2A-Tyr307 expression was observed. The enhanced p-AMPKα-Thr172 expression was inhibited by compound C, ryanodine (100 µM) and KN93 (10 µM). Simvastatin-induced p-PP2A-Tyr307 expression was suppressed by okadaic acid, compound C, ryanodine, KN93, phloridzin (1 mM), ouabain (10 µM), and in [glucose]o-free or [Na+]o-free conditions. Conclusions Simvastatin causes ryanodine-sensitive Ca2+ release which is important for AMPKα-Thr172 phosphorylation via Ca2+/CaMK II. AMPKα-Thr172 phosphorylation causes [glucose]o uptake (and an [ATP]i increase), closure of KATP channels, and phosphorylation of AMPKα-Thr172 and PP2A-Tyr307 resulted. Phosphorylation of PP2A-Tyr307 occurs at a site downstream of AMPKα-Thr172 phosphorylation.


Introduction
3-Hydroxy-3-methyl-glutaryl coenzyme A (HMG-CoA) reductase is a 97-kDa glycoprotein embedded in the endoplasmic reticulum [1] which is involved in the endogenous cholesterol biosynthesis in mammalian liver and intestine [2]. Pervious study of our group [3] has clearly illustrated the biochemical existence of extra-hepatic HMG-CoA reductase in human and porcine cardiovascular tissues, suggesting a physiological role of this enzyme in the cardiovascular system. HMG-CoA reductase inhibitors, commonly known as statins, have been shown to be an effective treatment of hypercholesterolemia and cardiovascular diseases via its cholesterol-lowering property and cholesterolindependent effects (pleiotropic effects) [3,4,5,6,7,8].
Regulation of vascular tone relies on complex cellular mechanisms as well as the opening and closing of various ion channels. Previous studies have demonstrated that statins can modify the activities of different ion channels in blood vessels including L-type Ca 2+ channel and BK Ca channel [3,9,10,11]. In addition to Ca 2+ channels and BK Ca channels, ATP-sensitive K + (K ATP ) channels are abundant in vascular tissues and K ATP channels are also important in regulating the vascular tone [12]. In rat isolated aorta, cerivastatin-induced a glibenclamide (a K ATP channel blocker)-sensitive aortic relaxation [13] and pravastatin reduced myocardial infact size through opening of mitochondrial K ATP channels in rabbit [14]. However, a recent study reported that simvastatin, but not pravastatin, inhibited pinacidil (a K ATP channel opener)-induced relaxation of pig's isolated coronary arteries suggesting that different statins have differential effects on K ATP channels of different cells/tissues [40].
Similar to other ion channels, the opening and closing of K ATP channels are modulated by multiple cell signaling mechanisms, such as phosphorylation by protein kinase A (PKA) [15], protein kinase C (PKC) [16] and cGMP-dependent protein kinase (PKG) [17]. In addition, the intracellular ATP level is an essential determinant of K ATP channel gatings. It is well-known that AMPactivated protein kinase (AMPK) serves as a 'metabolic master regulator' which is sensitive to changes of intracellular AMP/ATP ratio. Activation of AMPK results in suppression of intracellular energy-consuming pathways and generation of ATP i.e. an increase in cellular ATP level. In mouse isolated pancreatic islets, activation of AMPK by AICAR (an AMPK activator) potentiated insulin secretion by inhibiting K ATP channel openings [18]. Moreover, phenformin (another AMPK activator), inhibited K ATP channel openings in mouse aortic smooth muscle cells [19], highlighting the participation of AMPK activity in K ATP channel gatings in VSMC. Unfortunately, in various ex vivo studies (multicellular preparations), there is no consensus on the vascular effects mediated by AMPK activation as both contraction and relaxation were observed [20,21,22,23,24], and the underlying reason(s) for the discrepancy is unknown. Given the fact that statins promoted phosphorylation of AMPK in human and bovine endothelial cells [25], it is tempting to suggest that activation of AMPK by simvastatin could modulate vascular K ATP channel gatings and vascular reactivity.
Therefore, in this study we hypothesize that acute simvastatin could modulate vascular K ATP channel gatings and the simvastatin-mediated effects involve activation of AMPK signaling pathway. Thus, in this study, experiments were designed to evaluate the effects of acute simvastatin on vascular K ATP channel gatings of pig's coronary artery, and the participation of AMPK activation.

Animal and Human Ethics Statements
This investigation conformed to the Guide for the Care and Use of Laboratory Animals published by the US National Institute of Health (NIH Publication No. 85-23, revised 1996). The protocol was approved by the Animal Ethics Committee of the Chinese University of Hong Kong (Approval Number: 10/003/DRG). Permission prior to the collection of fresh pig's heart for research purposes was obtained from Sheung Shui Slaughterhouse (Hong Kong).
Fresh human left internal mammary arteries were the leftover obtained from patients with cardiovascular diseases undergoing coronary artery bypass grafting (CABG) procedures, and the use of human tissues for research purposes was approved by the Human Research Ethics Committee of the Chinese University of Hong Kong (CREC Ref. No. 2006.313). Written consents were obtained, prior to surgery, from patients voluntarily involved for the usage of tissues solely for research purposes. Patients had read and understood the patient information document provided, and the aims and methods of this study had been fully explained to them. Patients involved had given written informed consent (as outlined in PLOS consent form) to authors of this manuscript for publication of these data.

Isometric Tension Measurement
Fresh hearts were obtained from pigs (,35 kg) that were slaughtered in the morning of the experiment at a local slaughterhouse. The heart was immediately immersed in an icecold physiological salt solution. Segment of the left anterior descending (LAD) coronary artery (tertiary branch, O.D. ,500-800 mm) was dissected within an hour after the animal was slaughtered.
Fresh human left internal mammary arteries were bathed in an ice-cold physiological salt solution before transported to the laboratory from the operation theatre of the Prince of Wales Hospital (Hong Kong) within an hour. Fat and connective tissues were carefully removed under the dissecting stereo-microscope.
Arterial rings (porcine coronary artery and human internal mammary artery) (endothelium was removed using a blunted watch-maker forceps) were bathed in a 5-ml thermo-regulated wire myograph contained physiological salt solution with the composition (mM): NaCl 118.3, KCl 4.6, MgSO 4 1.2, NaHCO 3 25, KH 2 PO 4 1.2, CaCl 2 2.5, and glucose 11 (bubbled with 16%O 2 /5%CO 2 balanced with N 2 , pO 2 = ,100 mmHg). Rings (1 mm in length) were equilibrated under resting tension of 1 g [26,27] using two stainless steel wires (diameter ,100 mm), in the bath solution for 90 min. Resting tension was re-adjusted, if necessary, before commencing the experiments. The reason for choosing these tissues in this study is because previous reports demonstrated the existence of K ATP channels in these vascular tissues [28,29].

Enzymatic Dissociation of Myocytes
Porcine left anterior descending coronary artery myocytes and human left internal mammary artery myocytes were dissociated using collagenase and protease, as reported [3,13] for conventional whole-cell patch-clamp electrophysiology experiments.

Electrophysiological Measurement of K ATP Gatings
Conventional whole-cell, patch-clamp experiments were performed at room temperature (, 22uC) using single-cell, voltageclamp techniques (Axopatch 200B amplifier and Digidata 1200 A/D interface) (Axon Instruments, USA) with recording patch pipettes of 2-4 MV (when filled with internal pipette solution). Whole-cell recording configurations were used so as to maintain a ''pre-determined concentration'' of ATP (i.e. 1 mM) inside all the cells used during the recording of K ATP channels for a fair/accurate comparison of K ATP channel gatings of different cells (pig coronary artery and human internal mammary artery) in response to drug challenges. In addition, this mode of recording offers the convenience of a rapid delivery of drugs (e.g. simvastatin Na + , okadaic acid and rottlerin) into the cytosol of cells.

Determination of Cellular ATP Contents
ATP was extracted from cultured porcine coronary artery myocytes (before and after drug treatments) by trichloroacetic acid (final concentration, 0.5% vol./vol.). Trichloroacetic acid in the sample was then neutralized and diluted to a final concentration of 0.1% by adding Tris-acetate buffer (pH 7.75). The ATP content was analyzed using the ATP-dependent luciferin-luciferase bioluminescence assay (ENLITENH ATP assay system, Promega, USA).

Statistical Analysis
All data were obtained from at least 6 independent experiments. Statistical analysis was performed using Student's t test or ANOVA (one-way or two-way), where appropriate. A P value of ,0.05 was considered significant. Data are expressed as mean 6 S.EM.

Biochemical Existence of HMG-CoA Reductase and the Effects of Simvastatin and Simvastatin Na + on HMG-CoA Reductase Expression
The biochemical existence of HMG CoA reductase was determined in both human isolated left internal mammary artery and porcine isolated coronary artery. Porcine liver served as the positive control. Western blot results confirmed the biochemical existence of HMG-CoA reductase in both human and porcine vascular preparations. Beta actin was used as a loading control ( Figure 1A).
We then investigated the effects of simvastatin and simvastatin Na + in inhibiting HMG CoA reductase activity (i.e. phosphorylation of HMG CoA reductase) [29]. Neither simvastatin nor simvastatin Na + (10 mM, incubation # 30 min) altered the protein expression of p-HMG-CoA reductase-Ser 871 and HMG-CoA reductase in porcine isolated coronary artery ( Figure 1B).

Effects of Simvastatin on K ATP Channel Opener-induced Relaxation
To evaluate the involvement of K ATP channels, effects of simvastatin on K ATP channel opener-mediated vascular relaxation were examined. Cromakalim and pinacidil (both are K ATP channel openers) (10 nM to 10 mM) caused a glibenclamide (1 and 3 mM)-sensitive relaxation of U46619 (10 nM) pre-constricted coronary artery (endothelium-denuded) relaxation in a concentration-dependent manner (data not shown). Glibenclamide alone did not alter the basal tension and U46619-induced contraction. Simvastatin (3 and 10 mM), but not simvastatin Na + (1, 3 and 10 mM), significantly attenuated cromakalim-and pinacidilinduced relaxation (Figure 2A and B). Neither simvastatin nor simvastatin Na + altered the basal tension of the preparation.

Effects Simvastatin on K ATP Openings
In order to get a better understanding on the modulation of K ATP channels gatings by simvastatin, experiments were performed in single vascular myocytes. Cromakalim (10 mM) ( Figure 3A) and pinacidil (10 mM) (data not shown) significantly enhanced the recorded outward K + current amplitude which is inhibited by glibenclamide (a K ATP channel blocker), indicating that the recorded K + current is the genuine K ATP current.
In human internal mammary artery myocytes, neither simvastatin (1, 3 and 10 mM) nor simvastatin Na + (1, 3 and 10 mM) altered the basal K ATP channel gatings (data not shown). Interestingly, simvastatin caused a concentration-dependent inhibition of cromakalim (10 mM)-induced K ATP channel opening, with no apparent recovery after washout ( Figure 3B). However, simvastatin Na + (10 mM, applied either in external bath solution or included in the pipette solution) did not alter cromakalim (10 mM)induced K ATP opening (data not shown).
Due to the irregular/limited supply of human left internal mammary artery for research purposes, the following experiments were performed using porcine coronary artery myocytes. All drugs/inhibitors were tested against both cromakalim-and pinacidil-induced K ATP opening, however only representative figures of drug modulation of cromakalim-mediated responses were illustrated in the Figures.
The involvement of AMPK on cromakalim-and pinacidilinduced K ATP channel openings was examined. Similar to simvastatin (10 mM), AICAR (1 mM, an AMPK activator) attenuated cromakalim-and pinacidil-induced K ATP channel openings ( Figure 3D). However, AICAR (1 mM) did not alter the basal K ATP amplitude (data not shown).

Effects Simvastatin and AICAR on [Glucose] o Uptake
In order to establish the role of [glucose] o , effects of simvastatin and AICAR on [glucose] o uptake was determined. Simvastatin (10 mM) and AICAR (1 mM) caused a significant increase in [ 3 H]-2-deoxy-glucose uptake into coronary artery myocytes, and the ''enhanced'' [glucose] o uptake was eradicated by Compound C (10 mM) ( Figure 7A).

Effects of Simvastatin on [ATP] i Levels and LKB1 Activation
To confirm the generation of [ATP] i after [glucose] o uptake induced by simvastatin, cellular ATP level was estimated in response to drug challenges. AICAR caused a Compound C (10 mM)-sensitive increase in cellular ATP level of the arterial Figure 3. Effects of simvastatin on K ATP channel openings. (A) Effects of cromakalim (Crom., 10 mM) on whole-cell K ATP channel openings of single human internal mammary artery myocytes in the presence of glibenclamide (Glib., 3 mM) (n = 5 to 6). (B) Effects of cromakalim (Crom., 10 mM) on whole-cell K ATP channel openings of single human internal mammary artery myocytes with and without simvastatin (1, 3 and 10 mM). (C) Effects of simvastatin (10 mM) and glibenclamide (Glib., 3 mM) on cromakalim (Crom., 10 mM)-induced whole-cell K ATP channel openings of single porcine artery myocytes (in the presence of okadaic acid, 10 nM). Number of cells studied is indicated in parenthesis. *P,0.05, **P,0.01 and ***P,0.001 compared to controls. (D) Effects of cromakalim (Crom., 10 mM) on whole-cell K ATP channel openings of single porcine coronary artery myocytes in the presence of AICAR (1 mM). Number of cells studied is indicated in parenthesis. *P,0.05, **P,0.01 and ***P,0.001 compared to controls. doi:10.1371/journal.pone.0066404.g003 myocytes ( Figure 8A) and a time-dependent (2 to 30 min) increase in LKB1 activity (i.e. an increase in p-LKB1-Ser 428 /total LKB1 [33]) ( Figure 8B). However, simvastatin increased intracellular ATP level of the arterial myocytes with no apparent change in p-LKB1/total LKB1 ( Figure 8C).

Participation of Cytochrome P450 3A4
To elucidate the importance of cytochrome P450 (CYP450)mediated drug metabolism in mediating simvastatin-induced responses, effects of CYP450 3A4 inhibitor was examined. The biochemical existence of CYP450 3A4 protein was confirmed in porcine coronary artery and human internal mammary artery ( Figure 9A). Porcine liver served as the positive control. Ketoconazole (10 mM, a selective CYP450 3A4 inhibitor) failed to modify simvastatin (10 mM)-induced changes of AMPK and PP2A activities ( Figure 9B and C).

Discussion
In this study, acute simvastatin (membrane permeable) suppressed cromakalim-and pinacidil-induced relaxation of U46619 pre-constricted (endothelium-denuded) arteries with no effects on basal tension. In single myocytes of porcine coronary artery and human left internal mammary artery, simvastatin and AICAR inhibited cromakalim-and pinacidil-evoked K ATP channel openings with no apparent effect on basal K ATP channel gatings. Thus, a prerequisite opening of K ATP channels (by two structurally different K ATP channel openers cromakalim and pinacidil) is necessary for simvastatin and AICAR to demonstrate K ATP channel blocking properties. However, simvastatin Na + (membrane impermeable) did not alter the cromakalim2/ pinacidil-induced relaxation and the K ATP channel openings. Therefore, these results suggest that the lipophilic property of simvastatin is essential [3,34].
Acute application of HMG-CoA reductase inhibitors (pravastatin, atorvastatin and cerivastatin) elicited an endotheliumdependent relaxation of pre-constricted rat isolated aorta, and cerivastatin-induced relaxation was attenuated by glibenclamide and ouabain [13]. Activation of cardiac K ATP channels by statins has been reported [14,35,36]. However, a recent study [37] and our current study demonstrated that simvastatin inhibits pinacidilinduced relaxation in porcine isolated coronary artery. Furthermore, our result illustrate that simvastatin consistently suppressed, instead of enhanced, cromakalim-and pinacidil-induced K ATP channel openings of arterial myocytes of pig coronary artery and human left internal mammary artery. Taken together, these results clearly illustrate that simvastatin could alter vasodilatation via the inhibition of K ATP channels. AMPK (formerly termed HMG-CoA reductase kinase) is activated by an increase in [AMP/ATP] i ratio and a rise in [Ca 2+ ] i which signals an increase in energy demands [38,39]. Once activated, AMPK decreases ATP consumption and/or stimulates ATP production (e.g. via oxidative phosphorylation) and the [ATP] i level is thus restored [40]. In human umbilical vein endothelial cells (HUVECs), atorvastatin activated AMPK [25] whereas in mouse pancreatic islets b-cells, AICAR (an AMPK activator) inhibited K ATP openings [41]. Acute application of simvastatin significantly suppressed vasoconstriction of rat mesenteric resistance arteries via an AMPKa-phosphorylation-dependent mechanism [42]. In addition, AICAR activates AMPK via an increased phosphorylation, and phosphorylation (i.e. inactivation) of a known target for AMPK i.e. HMG-CoA reductase occurred [18,43,44]. However, in our study neither simvastatin nor simvastatin Na + (incubation # 30 min) altered the expression of HMG-CoA reductase and p-HMG-CoA reductase-Ser 871 (the inactivated isoform of HMG-CoA reductase) suggesting that AICAR and simvastatin are acting through different cellular mechanisms (see below). In rat's liver, AMPK is associated with HMG-CoA reductase [45]. In our study, simvastatin and AICAR  consistently increased p-AMPKa-Thr 172 expression (i.e. activation) [44], and suppressed cromakalim-and pinacidil-induced K ATP channel openings. Taken together, our results illustrate that acute simvastatin inhibits vascular K ATP channel openings probably via AMPKa-Thr 172 phosphorylation (i.e. activation).
The biochemical existence of cytochrome P450 (CYP450) 3A4 was demonstrated in porcine coronary artery and human left internal mammary artery, and the possible local enzymatic conversion of simvastatin into simvastatin Na + by CYP450 3A4 [61] was considered. However, ketoconazole (a selective CYP450 3A4 inhibitor) failed to modify simvastatin-induced increase in p-AMPKa-Thr 172 and p-PP2A-Tyr 307 expression refuted the possibility of local bio-transformation of simvastatin. Thus, our results strengthen the conclusion on the involvement of simvastatin, but not simvastatin Na + , in inhibiting vascular K ATP channel openings.
In conclusion, our results demonstrate that acute simvastatin caused phosphorylation of PP2A-Tyr 307 and AMPKa-Thr 172 , but not HMG-CoA reductase-Ser 871 , of porcine coronary artery. Ryanodine-sensitive Ca 2+ stores, but not [Ca 2+ ] o entry, play an obligatory role in simvastatin-elicited [Ca 2+ ] i increase and initiated the activation of Ca 2+ /CaMK II cascade which are essential for the subsequent AMPKa-Thr 172 phosphorylation (activation). Activation of AMPKa leads to [glucose] o uptake (and [ATP] i elevation resulted) with the participation of SGLT-1 and Na + /K + ATPase. An increase in [ATP] i levels not only closed the K ATP openers-induced channels openings but also provided the necessary phosphate groups for protein phosphorylation. Phosphorylation (inactivation) of PP2A-Tyr 307 probably occurs at a site downstream of AMPKa-Thr 172 phosphorylation ( Figure 10).