Beta Adrenergic Overstimulation Impaired Vascular Contractility via Actin-Cytoskeleton Disorganization in Rabbit Cerebral Artery

Background and Purpose Beta adrenergic overstimulation may increase the vascular damage and stroke. However, the underlying mechanisms of beta adrenergic overstimulation in cerebrovascular dysfunctions are not well known. We investigated the possible cerebrovascular dysfunction response to isoproterenol induced beta-adrenergic overstimulation (ISO) in rabbit cerebral arteries (CAs). Methods ISO was induced in six weeks aged male New Zealand white rabbit (0.8–1.0 kg) by 7-days isoproterenol injection (300 μg/kg/day). We investigated the alteration of protein expression in ISO treated CAs using 2DE proteomics and western blot analysis. Systemic properties of 2DE proteomics result were analyzed using bioinformatics software. ROS generation and following DNA damage were assessed to evaluate deteriorative effect of ISO on CAs. Intracellular Ca2+ level change and vascular contractile response to vasoactive drug, angiotensin II (Ang II), were assessed to evaluate functional alteration of ISO treated CAs. Ang II-induced ROS generation was assessed to evaluated involvement of ROS generation in CA contractility. Results Proteomic analysis revealed remarkably decreased expression of cytoskeleton organizing proteins (e.g. actin related protein 1A and 2, α-actin, capping protein Z beta, and vimentin) and anti-oxidative stress proteins (e.g. heat shock protein 9A and stress-induced-phosphoprotein 1) in ISO-CAs. As a cause of dysregulation of actin-cytoskeleton organization, we found decreased level of RhoA and ROCK1, which are major regulators of actin-cytoskeleton organization. As functional consequences of proteomic alteration, we found the decreased transient Ca2+ efflux and constriction response to angiotensin II and high K+ in ISO-CAs. ISO also increased basal ROS generation and induced oxidative damage in CA; however, it decreased the Ang II-induced ROS generation rate. These results indicate that ISO disrupted actin cytoskeleton proteome network through down-regulation of RhoA/ROCK1 proteins and increased oxidative damage, which consequently led to contractile dysfunction in CA.

Introduction b-adrenergic receptor (bAR) stimulation is a critical physiological mechanism for robust ''fight or flight response''. However, overstimulation of bAR cause pathological left ventricular hypertrophy (LVH), which is a potent, independent predictor of cardiovascular diseases including stroke, coronary heart disease and heart failure [1,2]. Compared with well established pathological event of bAR stimulation in heart, its effect on vasculature, especially cerebrovasculature, is still unknown.
Isoproterenol (ISO) is a synthetic catecholamine that is widely used for stimulation of all subtypes of bAR in cell [3] and animal model [4]. In the cultured cells, ISO-induced bAR stimulation activated ERK in cardiomyocytes [5] and astrocytes via PKA pathway [6]. In the rat aorta, 7 days of ISO treatment induced endothelial dysfunction and increased vasoconstriction [7]. In our previous studies, we demonstrated that ISO-bAR stimulation is associated with the modulation of Ca 2+ -activated K + , inward rectifier K + , and voltage-dependent K + channels in coronary arterial smooth muscle cells, which suggested functional modification of arterial smooth muscle cells during bAR stimulation [8,9,10]. We also found that ISO-bAR overstimulation disrupted the signaling of Ras/Raf/ERK cascades and highly increased activation of ERK in isoproterenol treated cerebral artery(CA) [4]. Since the Ras/Raf/ERK cascade is an important regulatory mechanism for vascular contractility, our previous findings suggested that bAR overstimulation is involved in cerebrovascular events [11,12,13]. However, functional consequences and responsible proteomic alteration of the ISO-bAR stimulation in cerebrovasculature were not evaluated. Therefore, we investigated the effect of bAR stimulation on cerebrovasculature using isoproterenol injected rabbit model. We tested whether bAR stimulation caused cerebrovascular damage then identified the proteomic alteration of CA and constructed protein interaction map of CA in bAR stimulation. Based on the proteomics data, we further demonstrated that bAR stimulation modified CA contractility through modulation of Ca 2+ mobility and ROS generation.

Ethics Statement
All experimental procedures were approved by the Institutional Review Board of Animals, Inje University College of Medicine (approval number: 2011-062). All surgery was performed under sodium pentobarbital anesthesia, and all efforts were made to minimize suffering.

Cell and tissue preparation
Enzymatic isolation of CA single smooth muscle cells (SMCs) was performed as previously described [8,9]. In detail, rabbit brains of Con and ISO model were isolated and placed in ice-cold (4uC) isolation normal tyrode (NT) solution containing 143 mM NaCl, 5.4 mM KCl, 1.8 mM CaCl 2 , 0.5 mM MgCl 2 , 5.5 mM glucose, and 5 mM HEPES (pH 7.4) adjusted with NaOH. The middle cerebral artery was dissected from the brain and disbranched. The isolated CA was then placed into Ca 2+ -free isolation solution containing 1.5 mg/ml papain, 1 mg/ml dithioerythreitol, and 1 mg/ml bovine serum albumin (BSA) for 10 min at 37uC, and then transferred into Ca 2+ -free isolation solution containing 1 mg/ml collagenase, 1 mg/ml hyaluronidase, and 1 mg/ml BSA for 8 min at 37uC. The enzyme-treated CA was washed three times in ice-cold isolation solution for 2 min. Finally, the CA was gently agitated using a polished glass pipette to obtain single SMCs.

Pressurized arterial experiment
To assess functional modifications of ISO-CAs, we assessed drug-specific contraction in response to high K + (KCl 60 mM) and angiotensin II (Ang II) in endothelium-denuded CA. Arterial diameter and drug-specific responses were assessed as previously described [9]. Briefly, the isolated middle cerebral artery (n = 4 in each group) was cannulated at both ends with micropipettes, secured with nylon monofilament suture and placed in a specially designed, custom-built chamber. The arteries were maintained in no-flow state and held at a constant intraluminal pressure of 60 mmHg. The extraluminal diameter of the artery was measured with a video edge detector (Crescent Electronics, Sandy, Utah, USA). High K + -induced arterial contraction was measured in 60 mM of KCl in intraluminal solution [16]. Ang II-induced arterial contraction was measured in dose dependent manner over concentration range from 1610 29 to 1610 24 M of Ang II in intraluminal solution. Percentage of contraction of each sample was compared to non-treated basal vessel diameter.

Angiotensin II-induced intracellular Ca 2+ release measurement
Ang II-induced intracellular Ca 2+ release rates were measured in the isolated SMCs of Con (n = 4) and ISO-CA (n = 4) using the membrane-permeant (acetoxymethyl ester) form of the Ca 2+sensitive fluorescent dye Fura 2 (Fura 2-AM) and photomultipliers as previously described [17]. Enzymatically isolated SMCs (n = 4 in each group) were loaded by incubation in 3 mM Fura 2-AM for 30 min at 37uC, and the extracellular Fura 2-AM (Invitrogen, USA) was then rinsed off with normal Tyrode solution. Monochromatic excitation light (355 nm) was delivered to the cell using a filter wheel (Life Science Resources, Cambridge, UK) via a liquid light guide and an oil-immersion objective lens (X40, NA 1.3; Nikon). The light emitted through an aperture slightly larger than the cell was measured simultaneously at 340 and 380 nm, and Ca 2+ concentration was estimated from the ratio of the fluorescence signals (340/380) obtained from the two photomultipliers (Life Science Resources) [17]. After 120 second stabilization, angiotensin II (1 mm/L)induced intracellular Ca 2+ level ([Ca 2+ ] i ) elevation were measured. Difference between basal (absence of Ang II)-to-peak (presence of Ang II) fluorescence ratio in each group was compared to the calculated quantitative group data. At the same time, fluorescence images of calcium transient in Fluo-4AM stained (Invitrogen, USA, 1 mM, incubation in 30 min at 37uC) Con and ISO SMCs were acquired using confocal microscope

Ang II-induced ROS generation assay
The Ang II-induced ROS production was assessed using reactive oxygen species (ROS) indicator CM-H 2 DCF-DA (Invitrogen, USA). Fluorescence excitation and emission wave lengths were ,492-495 and 517-527 nm, respectively. Enzy- ROS levels were defined as the ratio between the mean fluorescence of cells before and after the treatment with Ang II. Superoxide production assay We further measured the Ang II-independent ROS generation in CA-SMCs. NADPH-dependent O 2 production by CA homogenates was measured using SOD-inhibitable cytochrome c reduction assay as previously described [18]. ISO-(n = 6) and control-CA (n = 6) homogenates (final concentration 1 mg/mL) were distributed in 96-well flat-bottom plates (final volume 200 mL/well). Cytochrome c (500 mmol/L) and NADPH (100 mmol/L) were added in the presence or absence of superoxide dismutase (SOD, 200 U/mL) and incubated at room temperature for 30 minutes. Cytochrome c reduction was measured by reading absorbance at 550 nm on a microplate reader. O 2 production in nmol/mg protein was calculated from the difference between absorbance with or without SOD and the extinction coefficient, DE 550 = 2.1610 4 M 21 cm 21 , for change of ferricytochrome c to ferrocytochrome c. Superoxide production was expressed in units of 'O2-nmol/mg protein'.

Lipid peroxidation assay
Lipid peroxidation or malondialdehyde (MDA) assay is a wellestablished method for measuring oxidative cellular injury in cells and tissues. MDA concentration in Con and ISO-CAs were evaluated using lipid peroxidation assay kit (FR12, Oxford biomedical, USA) [19].

Two-dimensional gel electrophoresis (2-DE) proteome analysis
Proteomic differences between normal control CA and ISO-CA were assessed by 2-DE proteome analysis and MALDI-TOF MS analysis as previously described [21]. Collected normal (n = 3) and ISO-CA (n = 3) were dissolved by repeated vortexing and sonication in ice-cold lysis buffer (7 mol/L urea, 2 mol/L thiourea, 4% CHAPS, 40 mM Tris base, 1% DTT, 0.5% IPG buffer, 0.5% Triton X-114, and protease inhibitor cocktail) for 1 h. The CA protein concentration was assayed using a 2D Quant kit (GE Healthcare) following manufacturer's instruction, and two-dimensional gel electrophoresis (2-DE) and proteome analysis were performed. Each sample was run in duplicate. 13 cm-long dehydrated and immobilized pH gradient (IPG) strips with nonlinear pH range from 3 to 10 were rehydrated overnight in rehydration tray with 250 ml of Destreak TM rehydration solution (GE Healthcare) containing 2% IPG buffer (v/v). Subsequently, 50 mg of the soluble CA proteins in total 100 ml sample solution were loaded by cup loading method. Isoelectric focusing was carried out at 80,000 V/h at 20uC as follows: 500 V for 1 h, 1,000 V for 1 h, and finally gradual voltage increase from 8,000 V  to 80,000 V over an hour. After focusing, the IPG strips were placed in 5 ml of an equilibration solution (50 mM Tris-HCl, pH 8.8, containing 6 M urea, 30% glycerol, 2% SDS, and bromophenol blue) that contained 1% DTT (v/v) during the first equilibration step and 2.5% iodoacetamide (v/v) during the second equilibration step (15 min per equilibration step). The 2D separation was performed using the SE600 system (Amersham). The IPG strips were loaded onto a 12.5% SDS-polyacrylamide gel and sealed with low melting point agarose. 4 L running buffer (25 mM Tris, 192 mM glycine, and 3.5 mM SDS at pH 8.3) was filled in the electrophoresis chamber, and 80 volt was applied for 30 min in the first step and 200 volt for around 4 hr in the second step, until the probe dye reached 1 mm distance from the bottom of the gel. The gels were then stained with silver nitrate.
Silver-stained gels were scanned on a flatbed scanner (Umax PowerLook 1100; Fremont, CA, USA), and the digitized images were analyzed using automated image analysis algorithm software (ImageMaster 2D Platinum version 5.0, Amersham Biosciences). Spots showing the same distribution pattern in all gels were selected for further analysis. To increase the confidence level, we filtered the detected protein spots with class gap values over 0 and selected spots that showed a .1.5-fold change in expression and Student's t-test P-value of ,0.05 compared to the control. Protein quantification was calculated using the percent volumes (% Vol) of identified proteins for control and isoproterenol stimulated cerebral artery as shown in the equation below.
where Vol s is the volume of spot S in a gel containing n spots.

Protein identification using MALDI-TOF
Selected spots were enzymatically digested by trypsin and analyzed using matrix-assisted laser desorption ionization time of flight (MALDI-TOF). Selected protein spots were excised from each gel and destained with 30 mM potassium ferricyanide and 100 mM sodium thiosulfate. After washing with 50% acetonitrile (ACN), gel fragments were dried in a vacuum centrifuge. The protein in gel was digested by 0.5 mg of sequencing-grade trypsin (Promega, Southampton, UK) in 20 ml of 25 mM ammonium bicarbonate at 37uC overnight. Digested peptides were extracted with 0.5% trifluoroacetic acid (TFA)/50% ACN solution, and then desalted using ZipTip C18 (Millipore, Bedford, MA) tip. Peptides were eluted directly onto MALDI target by a-cyano-4hydroxy-cinnamic acid (CHCA) matrix solution (10 mg/ml CHCA in 0.5% TFA/50% acetonitrile (1:1, (v/v)). All mass spectra were acquired at a reflection mode by a 4700 Proteomics Analyzer (Applied Biosystems, Framingham, MA). The MS spectra were analyzed using Peak Explorer TM 3.0 (Applied Biosystems) software. Resulting data were analyzed by GPS Explorer TM 3.5 (Applied Biosystems) software. The proteins were identified by searching Mammalia of the National Center for Biotechnology Information (NCBI) protein databases using MASCOT 2.0 search algorithm (Matrix Science, London).

Western blotting
Randomly selected proteome analysis results were confirmed by Western blotting analysis as previously described [22]. Briefly, the protein concentration of each sample (n = 3 in each group) was determined using the bicinchoninic acid (BCA) protein assay. Samples containing 20 mg of total protein were subjected to 12% SDS-polyacrylamide gel electrophoresis (PAGE) and then electrophoretically transferred to a polyvinylidene difluoride (PVDF) membrane. After blocking with 5% non-fat dried milk in Trisbuffered saline/Tween-20 (TBST) for 1 h at room temperature, the membrane was incubated for 2 h with randomly selected primary antibodies including anti-actin related protein (ACTR1A,

Functional annotation and associated network analysis
In order to understand the physiological significance of proteomic alterations in ISO-CAs, we categorized and annotated the identified proteins using orthologous groups of proteins (COGs) algorithm (http://www.ncbi.nlm.nih.gov/COG/grace/ fiew.cgi) and Protein ANalysis THrough Evolutionary Relationships (PANTHER) classification algorithm (http://pantherbeta.ai. sri.com/help/PANTHERhelp.jsp) [23]. We analyzed proteinprotein interaction and functional associations between identified proteins and constructed a functional network using the online STRING 8.0 database (http://string.embl.de) [24]. The constructed protein network was visualized with Cytoscape software (Version 2.6.3).

Statistical analysis
All results were expressed as mean 6 standard error (SE). Differences between control and ISO-CAs were analyzed using two-tailed Student's t-test. Significance of dose-dependent Ang II induced vascular contraction between control and ISO-CAs was tested by two-way ANOVA analysis using Origin Pro 8.0. P-values ,0.05 were considered statistically significant.

Effect of ISO-bAR stimulation on the heart and hemodynamics
After 7 days of daily ISO injection, hearts of ISO treated animals were significantly enlarged than normal group animals ( Figure S1A). The heart weight and heart to body weight ratio of ISO group were also significantly increased than those of normal group ( Figure S1B) indicating successful ISO-bAR stimulation in the model animal. However, systolic, diastolic, mean arterial pressure and heart beat rate were not significantly altered by ISO treatment (Table 1) indicating there was no hemodynamic effect of ISO-bAR stimulation on cerebrovasculature. Unlike to hypertrophied heart, there is no significant alteration of arterial lumen and thickness between control and ISO treated cerebral arteries ( Figure S1C). In addition, the ISO injection did not alter expression of angiotensin II type 1 (AT1R) and type 2 (AT2R) receptors itself ( Figure S1D).

Proteomic alteration in ISO-CAs
Protein spots (981.0661.0) were detected from scanned 2-DE gel images. Major spot dispersions were observed over a pH range of 4-9 and a molecular weight range of 10-100 kDa. The expression patterns of CA proteins in normal and ISO-rabbits are shown in Figure 1A. We identified differentially expressed 32 proteins including 2 up-and 30 down-regulated proteins using MALDI-TOF MS analysis ( Figure S2). Detailed MS information of identified proteins was listed in supporting information Table S1 and Figures S3 and S4. The identified proteins were divided into 11 groups based on COG category and their expressional changes are listed in Table 2. Through annotation and categorization of identified proteins, we found that modulated proteins were widely associated with regulation of cytoskeletons ( Figure 1C and D). Specifically, the expression of actin gamma 2 (ACTA3, 54%), actinrelated protein 1 (ACTR1A, 50%), actin-related protein 2 (ACTR2, 47%), a-actin (ACTC1, 37%), capping protein muscle Z-line (CAPZB, 55%), vimentin (VIM, 36%) and serine/threonine-protein phosphatase PP1-beta catalytic subunit (PPP-1B) decreased significantly in ISO-CAs ( Figure 1B). Additionally, expression of several neuroprotective chaperones and protein maturation elements, such as heat shock protein 9A (mortalin, 42%) and stress-inducedphosphoprotein 1 (STI1, 55%) were markedly down-regulated in ISO-CAs (Table 2 and Figure 1B).

Validation of Proteomic result
Western blotting was used to validate the changes in the expression of cytoskeletal proteins, which are closely related to CA contraction and dilation. The expression of cytoskeletal components and regulatory proteins, including the ACTR1A, ACTR2, a-actin, VIM and PPP-1B, significantly decreased in the ISO-CAs. In addition, the protein expression levels of NADH dehydrogenase and glutathione-S-transferase were confirmed to be increased in the ISO-CAs (Figure 2A and B). Decreased a-actin level was further validated in ISO-CAs by immunohistochemistry ( Figure 2C).

Functional association of altered proteins
In the primary protein-protein interaction search, we constructed three discontinuous clusters consisting of 23 protein nodes with 26 interactions among the nodes; the 9 identified proteins did not have known direct interaction with others ( Figure 3A). Subsequently, to extend the interacting partner search, we found 16 predicted interaction proteins that linked with 23 primary protein nodes, which included two important actin cytoskeleton regulatory proteins, RhoA and ROCK1. Furthermore, protein nodes in the network were marked in four different colors based on their major cellular functions: cytoskeletal and contractile regulation, energy production and conversion, chaperone and cellular signaling, and mitochondrial electron transfer chain. Consequently, we constructed a simple, non-redundant protein network consisting of 48 protein nodes with 94 interactions that may be helpful in understanding proteome-based systemic changes of CA in ISO. Western blotting and immunohistochemistry confirmed the decreased expression of RhoA and ROCK1 in ISO-CAs ( Figure 4A and B).

Superoxide production and DNA damage
To evaluate if ISO-bAR stimulation induces oxidative damage in cerebrovasculature . We measured ROS production, lipid peroxidation and DNA damage in the cerebral arterial SMCs of both group. In the presence of NADPH, O 2 production was significantly increased in the ISO-CAs group (7.560.85 O 2nmol/mg protein) than in the control group (3.660.40 O 2 -nmol/ mg protein, n = 3, p,0.05) ( Figure 5A). Increased O 2 production led directly to lipid oxidation ( Figure 5B) and severe DNA damage in ISO-CAs ( Figure 5C). Cells containing severely damaged DNA (class IV) were significantly more frequent in the ISO group (24.761.3%) than in the control group (9.563.2%) ( Figure 5D).

Ang II-induced calcium release and arterial contractility alteration
Ang II-induced vascular contraction experiments were applied to test whether proteomic alterations actually lead to modification of vascular contractile response. As a result of the experiments, high K + and Ang II-induced vascular contraction was significantly impaired in ISO-CAs compared to the control ( Figure 6A and B, Figure S5A). As for the causes of contractile dysfunction, we found that ISO treatment significantly reduced Ang II-induced [Ca 2+ ] i transient peak (Figure 6C and D) and prolonged [Ca 2+ ] i elevation ( Figure 6E and F and Figure S5B) and ROS generation rate ( Figure 6G and Figure S5C).

Discussion
Maintenance of normal blood circulation in the cerebral and cardiovascular systems is essential for life. Distorted cerebral homeostasis may aggravate the risk of many life-threatening neurodegenerative events including stroke. The bAR overstimulation-induced cardiac hypertrophy is believed to potentiate cerebral damage even in the absence of clinical symptoms like as hypertension [25,26,27,28,29,30], suggesting bAR overstimulation may cause cerebrovascular damage. In our model, ISO injectioninduced cardiac hypertrophy without changes of LV systolic pressure and LV end-diastolic pressure indicated successful induction of bAR overstimulation in the model animal. Also, we were able to exclude hemodynamic effect on cerebral artery in the model ( Figure S1 and Table 1) [15,31,32]. The ISO injection did not alter expression of angiotensin II type 1 (AT1R) and type 2 (AT2R) receptors ( Figure S1D).

Remodeled cytoskeletal proteome network in ISO-CAs
As a core finding of this study, we discovered that ISO significantly remodeled cytoskeletal proteome network, which disrupts vascular responses to Ang II. Major proteomic alteration in ISO-CAs is shown by remarkable down-regulation of cytoskeletal proteins (Table. 2 and Figure 2). This result was in agreement with our previous study that PKA activity and Ras/Raf expressions were significantly decreased in ISO-CAs [4]. Because Ras/Raf/ERK signaling cascade is essential for actin-base cytoskeleton organization, decreased level of those proteins in ISO-CAs interrupted actin cytoskeleton network [11,12,13]. As a reliable cause of actin-cytoskeletal disorganization, we found a decreased protein expression level of RhoA and ROCK1 (Figure 4). RhoA and ROCK1 play an essential role in the actin-cytoskeleton organization and smooth muscle contraction via phosphorylation of myosin light chain [33,34]. Thus, our result suggested that ISO stimulation increased ERK activation [4,5], which oppositely suppressed ROCK1 activity [35,36] and led to disorganization of actin-cytoskeleton. In addition, decreased level of vimentin [37] and moesin [38] may be implicated with altered activation of ERK and ROCK1 in ISO-CAs.
Cytoskeleton structure and its components are fundamentally important for maintaining cell shape and integrity. Increasing number of experiments suggest their dynamic role in various biological processes [39,40,41]. In the vascular system, contractility of SMCs is widely regulated by cytoskeletal proteome network [42]. Since actin is the major component of this cytoskeletal network, down-regulation of a-actin and actin-related proteins could lead to dysregulation of cytoskeletal network organization of ISO-CAs. Actin is the most essential and fundamental protein unit in terms of cellular structure and has great functional significance in the regulation of contractility in various tissue and muscles. In CAs, dynamic actin cytoskeleton regulates arterial diameter in response to changes in intravascular pressure through the polymerization of monomeric globular (G-) actin into filamentous (F-) actin [40]. Polymerization of the actin cytoskeleton is regulated by the Wiskott2Aldrich syndrome protein (WASP) and the Arp2/3 complex [43]. Furthermore, WASP and Arp2/3 activities are implicated with activities of moesin [44] and coronin 1B [45]. Therefore, down-regulated expression of a-actin, ACTR2 (or Arp2), moesin and coronin 1B suggests that ISO-stimulation could disrupt actin-based polymerization and cytoskeleton organization in CAs.
To evaluate functional relevance of those actin cytoskeletal protein modifications, we investigated the Ang-II-induced Ca 2+ regulation and contractile response because Ang-II is the most well known vasoconstrictor. Ang II-induced ROS generation and ERK signaling has important role in the Ang II-induced vascular contraction [46,47]. Actin cytoskeletal network also plays an important role in regulating Ang II-induced Ca 2+ release from internal stores and Ca 2+ influx [41] and L-type Ca 2+ channel current [48]. In agreement with these studies, we found that the disruption of cytoskeletal network in ISO-CAs led to decreased Ang II-induced cellular Ca 2+ elevation and contraction ( Figure 5). As well as in Ca 2+ signaling, actin-cytoskeleton plays a pivotal role in Ang II-induced ROS generation, which is essential for Ang IIinduced vascular contraction [47,49]. Disruption of actin cytoskeleton using cytochalasin B significantly reduced Ang II-induced ROS generation [49]. Similarly, Ang II-induced ROS generation was significantly decreased with disruption of actin-cytoskeleton in our result ( Figure 6G). These results suggest that disruption of actin cytoskeletal network interrupts Ang-II mediated intracellular Ca 2+ homeostasis, ROS generation and vascular contraction in ISO-CAs.

Increased oxidative stress in ISO-CA with loss of antioxidative proteins
Although ISO treatment blunted Ang II-ROS generation rate, it increased the basal ROS level and oxidative stress ( Figure 5). Similarly, several recent findings demonstrated that ISO stimulation increased reactive oxygen species (ROS) production through bAR in HEK293 cell [50], rat cardiac myocyte [51], rat aorta [15] and DNA damage of rat cardiac myocyte [52]. Our results showed the sensitivity of ISO-CAs to DNA damage. The evidence for accelerated oxidative stress was provided by our proteomic analysis result, which showed significantly decreased expression of several cytoprotective chaperones and protein maturation elements. In particular, decreased levels of HSP9A (mortalin) and stress-induced-phosphoprotein 1(STI1), which are cytoprotective proteins, are deleterious to the cellular anti-oxidative mechanism.
Mortalin is a member of the essential mitochondrial chaperone (heat shock protein 70) family and has a cellular protective role against various oxidative stresses through suppression of ROS production [53] and accumulation [54]. Even though the mechanism through which mortalin suppresses ROS production is unclear, mortalin may stabilize cytochrome c and other components of the electron transport chain (ETC) and thereby suppress mitochondrial ROS production [53]. The STI1, also known as Hsp70/Hsp90-organizing protein (HOP), is a linker of Hsp70 and Hsp90 and regulator of linked chaperones activities [55]. Recent study demonstrated that it has a role in neuroprotection [56] and regulates the activity of superoxide dismutase for ROS scavenging [57]. Thus, down-regulated expression of mortalin and STI1 may cause deleterious oxidative stress on DNA in ISO-CAs. Beside that, Hsp90, which is regulated by HOP, is essential for binding of Raf-1-Ras complex and regulating their activities. Down-regulation of HOP is therefore a significant factor in decreased Ras/Raf activities in ISO-CAs [58].
In addition to biological results, systemic analysis of proteomic data helped us to understand integrative biological significance of each altered proteins and to comprehend complicated interactions among the proteins. Since proteins rarely act alone but rather act in concert with other proteins to constitute a biological pathway [59], it is important to analyze interaction and functional clustering of each altered proteins in ISO-CAs for understanding cerebral arterial dysfunction in bAR overstimulation.

Conclusion
The present study demonstrated that bAR overstimulation increased oxidative stress by damaging anti-oxidative proteins and impaired contractile response of CA. As a possible mechanism of this abnormality, our results suggested that bAR overstimulation disrupted actin cytoskeleton proteome network through downregulation of RhoA/ROCK1 proteins and increased oxidative damage, which consequently led to contractile dysfunction in CA (Figure 7). Possible involvement of cytoskeletal disorganization in cerebrovascular dysfunction may give a new insight into understanding cerebral damage after bAR overstimulation and therapeutic intervention of bAR overstimulation induced cerebrovascular damage. Figure S1 The effect of beta adrenergic (bAR) overstimulation in heart and angiotensin 2 receptors in cerebral artery. A. Images of longitudinal sectioned heart of control (left) and isoproterenol injected (right) rabbit. B. Comparative histogram of body weight (BW), heart weight (HW) and HW/BW of control and isoproterenol treated rabbits. C. Images of Hematoxylin and Eosin stained cerebral arteries from control and ISO treated rabbits. D. Western blot analysis of angiotensin 2 type 1 (AT1R) and type 2 (AT2R) receptors in control and ISO-CAs (n = 5 in each group, Student's t-test *P,0.05 vs. control). (DOC)