Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Stiff Substrates Enhance Endothelial Oxidative Stress in Response to Protein Kinase C Activation

Stiff Substrates Enhance Endothelial Oxidative Stress in Response to Protein Kinase C Activation Hindawi Applied Bionics and Biomechanics Volume 2019, Article ID 6578492, 14 pages https://doi.org/10.1155/2019/6578492 Research Article Stiff Substrates Enhance Endothelial Oxidative Stress in Response to Protein Kinase C Activation 1 2 1,2 Rebecca Lownes Urbano, Swathi Swaminathan, and Alisa Morss Clyne Mechanical Engineering and Mechanics, Drexel University, Philadelphia, PA, USA Biomedical Engineering, Science and Health Systems, Drexel University, Philadelphia, PA, USA Correspondence should be addressed to Alisa Morss Clyne; asm67@drexel.edu Received 1 October 2018; Revised 28 January 2019; Accepted 19 February 2019; Published 14 April 2019 Academic Editor: Estefanía Peña Copyright © 2019 Rebecca Lownes Urbano et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Arterial stiffness, which increases with aging and hypertension, is an independent cardiovascular risk factor. While stiffer substrates are known to affect single endothelial cell morphology and migration, the effect of substrate stiffness on endothelial monolayer function is less understood. The objective of this study was to determine if substrate stiffness increased endothelial monolayer reactive oxygen species (ROS) in response to protein kinase C (PKC) activation and if this oxidative stress then impacted adherens junction integrity. Porcine aortic endothelial cells were cultured on varied stiffness polyacrylamide gels and treated with phorbol 12-myristate 13-acetate (PMA), which stimulates PKC and ROS without increasing actinomyosin contractility. PMA-treated endothelial cells on stiffer substrates increased ROS and adherens junction loss without increased contractility. ROS scavengers abrogated PMA effects on cell-cell junctions, with a more profound effect in cells on stiffer substrates. Finally, endothelial cells in aortae from elastin haploinsufficient mice (Eln+/-), which were stiffer than aortae from wild-type mice, showed decreased VE-cadherin colocalization with peripheral actin following PMA treatment. These data suggest that oxidative stress may be enhanced in endothelial cells in stiffer vessels, which could contribute to the association between arterial stiffness and cardiovascular disease. 1. Introduction (Eln+/-) mice had enhanced angiotensin-induced vasocon- striction and impaired endothelium-dependent vasodilation Due to the highly mechanical nature of the cardiovascular [13], although aortae from these same animals do not [16]. system, cardiovascular disease has long been accepted as both In human subjects, endothelial flow-mediated vasodilation was inversely correlated with aortic stiffness [17, 18]. Thus, a biomechanical and biochemical disease. Arterial stiffness, which increases with hypertension and aging among others, arterial stiffness alone may contribute to cardiovascular risk is an independent predictor of cardiovascular risk [1–3]. by altering critical endothelial functions. Arteries reversibly stiffen when smooth muscle cells contract However, cardiovascular risk factors rarely occur in isola- and irreversibly stiffen as elastin is degraded and collagen tion but rather cluster in certain individuals. Little is known increases [4–9]. Stiff arteries have long been known to con- about how arterial stiffness interacts with other cardiovascu- tribute to cardiovascular mortality by increasing cardiac lar risk factors such as diabetes and inflammation. Both afterload [10]; more recently, stiff arteries have also been hyperglycemia and inflammatory cytokines such as tumor shown to contribute to endothelial dysfunction, an initiat- necrosis factor-α (TNF-α) increase endothelial cell oxidative ing step in atherosclerosis [11–13]. In vitro, endothelial stress due to increased production and decreased scavenging monolayers on stiff polyacrylamide (PA) gels were more of reactive oxygen species (ROS). Elevated ROS have been permeable [12, 14, 15]. In animal models, endothelial per- implicated in both hypertension and atherosclerosis patho- meability was elevated in stiffened aortae from older mice genesis [19]. ROS are a family of highly reactive oxygen- [12]. Mesenteric arteries from elastin haploinsufficient containing molecules, including superoxide (O ), hydrogen 2 2 Applied Bionics and Biomechanics aorta was cut into four segments—two thoracic segments and peroxide (H O ), hydroxyl radical ( OH), and peroxynitrite 2 2 (ONOO ), which play an important role in many signaling two abdominal segments—producing four samples per aorta. pathways, such as cell proliferation, survival, and metabolism Samples were carefully mounted, the endothelium facing up, [20]. Superoxide is produced by the mitochondrial electron on a coverslip using Loctite 401 medical grade adhesive transport chain, as well as through protein kinase C- (PKC-) (Henkel) and submersed in PBS. The endothelium was induced NADPH oxidase upregulation and activation removed by gentle scraping with a cotton-tip applicator, [21–23]. NADPH oxidase assembly at the cell membrane based on a published protocol [31]. Subendothelial stiffness requires Rac, which is enhanced by substrate stiffness [24]. was determined by AFM using precalibrated cantilevers The effect of arterial stiffness on endothelial ROS produc- (spring constants between 0.10 and 0.17 N/m) with 10 μm tion in response to an external stimulus such as hyperglyce- spherical tips. Between three and nine indentations were mia or TNF-α has not yet been investigated. However, made at different locations along each sample. The force- these risk factors also activate many other endothelial cell sig- indentation curve for each indentation was fit to the Hertz naling pathways. Therefore, to isolate substrate stiffness model down to 200 nm indentation using a custom effects on endothelial ROS production, we used phorbol MATLAB code to produce a stiffness value [32]. Subendothe- 12-myristate 13-acetate (PMA), a PKC activator widely lial stiffness was calculated as the average of the individual used to stimulate ROS production in vitro. In previous stiffness values of each sample. studies, PMA treatment increased endothelial monolayer permeability but did not increase actinomyosin contractility, 2.3. Cell Culture and Polyacrylamide (PA) Gel Sample as measured by silicon substrate wrinkling, myosin light chain Preparation. Primary porcine aortic endothelial cells (PAEC) (MLC) phosphorylation, or MLC kinase activation [25, 26]. were isolated by the collagenase dispersion method and cul- Phorbol esters may instead induce barrier loss through tured in low glucose Dulbecco’s modified Eagle’s medium intermediate filament or actin cytoskeleton reorganization (DMEM, Corning) supplemented with 5% fetal bovine [27–29]. Thus, PMA enables investigation of substrate serum (FBS, HyClone), 1% glutamine, and 1% penicillin- stiffness effects on ROS production without stimulating streptomycin (Invitrogen). Cells were used up to passage 9. actinomyosin contractility. 6, 14, or 29 kPa was selected for the PA gel stiffnesses We hypothesized that stiff substrates would increase based on the subendothelial stiffnesses measured in WT endothelial ROS in response to PMA, resulting in actin fiber and Eln+/- mouse aorta (Figure 1). PA gels were prepared formation and cell-cell junction loss. We used varied stiffness following well-established protocols [33, 34]. Briefly, a bot- PA gels to study PMA-induced endothelial ROS, actin fiber tom coverslip was made hydrophilic by consecutive incuba- formation, and adherens junction loss. Abdominal aortae tions with 0.1 M sodium hydroxide (NaOH, Sigma-Aldrich), from wild-type (WT) and Eln+/- mice were treated with 3-aminopropyltrimethoxysiliane (3-APTES, Sigma-Aldrich), PMA ex vivo and imaged en face. We now show that sub- and 0.5% glutaraldehyde (Electron Microscopy Sciences). A strate stiffness enhances PMA-induced oxidative stress in top coverslip was made hydrophobic by applying SurfaSil endothelial monolayers in vitro and alters actin fiber reorga- (1,7-dichloro-octamethyltetrasiloxane, Thermo Scientific). nization and adherens junction morphology both in vitro A solution containing varying amounts of 40% acrylamide and ex vivo. and 2% bisacrylamide (Bio-Rad) was prepared based on the desired gel stiffness (Table 1). Ammonium persulfate (Bio-Rad) and tetramethylethylenediamine (TEMED, Bio- 2. Materials and Methods Rad) were added to the acrylamide/bisacrylamide solution 2.1. Animals. All experiments were performed according to to achieve final concentrations of 0.1% w/v and 0.3% v/v, respectively, initiating gel polymerization. Polymerizing gel protocols approved by the Drexel University College of Medicine Animal Studies Committee. Eln+/- mice were solution was added to the bottom coverslip, and the top generated as previously described [30]. 8-12-week-old coverslip was quickly inverted onto the polymerizing gel to create a flat surface. After gel formation, the top cover- WT and Eln+/- mice of both sexes, backcrossed several generations into the C57BL/6 background (Charles River), slip was removed. Elastic modulus was confirmed by AFM. To make the surface adhesive to cells, the gel was were used. All mice were genotyped to confirm elastin het- erozygosity, and decreased elastin lamellae thickness was UV-activated using sulfo-SANPAH (Thermo Fisher) in confirmed in select animals by immunohistochemistry. dimethyl sulfoxide (DMSO, Fisher Scientific) and 50 mM HEPES buffer and then incubated with 100 μg/mL type I Mice were provided access to food and water ad libitum ° ° at 22 C and a 12-hour light/dark cycle. collagen (BD Biosciences) at 37 C for 3 hours at room temperature or at 4 C overnight. The collagen-coated gel 2.2. Atomic Force Microscopy (AFM). AFM was used to quan- was rinsed in sterile phosphate-buffered saline (PBS) and tify aortic stiffness in wild-type (WT) and Eln+/- mouse aor- UV-sterilized prior to cell seeding. tae. The aorta was dissected and transferred to ice-cold PAEC were seeded on collagen-coated PA gels in phenol HEPES buffer (140 mM NaCl, 5 mM KCl, 1 mM CaCl , red-free DMEM and cultured to confluence for three days in 1.2 mM MgSO , 1.2 mM Na HPO , 10 mM HEPES, 10 mM a growth medium. Cells were then serum-starved overnight 4 2 4 sodium acetate, and 5 mM glucose, pH 7.4). Excess tissue in phenol red-free DMEM containing 1% FBS, 1% glutamine, was cleaned from the outside of the vessel, and the vessel and 1% penicillin-streptomycin. After serum starvation, cells was cut open longitudinally to expose the endothelium. Each were left untreated or treated with 1 μM PMA for varying Applied Bionics and Biomechanics 3 Thoracic Abdominal (a) (b) (c) Figure 1: Subendothelial stiffness increased in the thoracic and abdominal aortae of Eln+/- mice. (a) Longitudinally dissected mouse aorta opened to expose the endothelial surface. (b, top) Unscraped aorta showing the intact endothelium via β-catenin (green), cell structure using actin (red) and nuclei (blue), and the subendothelial matrix using collagen IV (white). Artery is shown en face. (b, bottom) Scraped aorta showed that the endothelium was removed since no β-catenin (green) was observed. The subendothelial matrix (collagen IV, white) remained intact and contiguous both en face and in cross section (smaller images). Scale bar = 50 μm. (c) Subendothelial stiffness of aortae from WT and Eln+/- mice. Thoracic and abdominal aortic sections were indented by AFM using a silicon nitride cantilever with a 10 μm spherical tip to measure subendothelial stiffness ( p <0 01 and p <0 05 by Student’s t-test). Three aortae were tested for each condition. Table 1: Acrylamide and bisacrylamide concentrations used to durations. In some cases, endothelial monolayers were pre- create varying elastic modulus PA gels. treated with ROS scavengers (4 mM N-acetyl cysteine or 50 mM sodium pyruvate, Sigma) for 1 hour prior to PMA. Acrylamide Bisacrylamide Elastic modulus 7.5% 0.05% 6 kPa 2.4. ROS Assay. ROS were measured using 5-(and-6)- 10% 0.1% 14 kPa ′ ′ chloromethyl-2 ,7 -dichlorodihydrofluorescein diacetate (CM-H DCFDA), which passively diffuses into cells where 10% 0.3% 29 kPa it is cleaved by intracellular esterases and then oxidized by Scraped Unscraped 4 Applied Bionics and Biomechanics ROS to yield a fluorescent adduct. 100 μM tert-butyl collected in prechilled Eppendorf tubes, and centrifuged at hydroperoxide (tBHP), which produces intracellular hydro- 4 C for 15 minutes at 13,000 rpm. The supernatant was col- gen peroxide, was the positive control. After PMA or tBHP lected, and the protein concentration was determined by treatment, samples were rinsed with warmed HBSS buffer BCA assay (Thermo Fisher). PKC activity was quantified in (0.137 M NaCl, 5.4 mM KCl, 0.25 mM Na HPO , 5.6 mM control or treated cell lysates using an ELISA-based PKC 2 4 glucose, 0.44 mM KH PO , 1.3 mM CaCl , 1.0 mM MgSO , kinase assay (Enzo) as per the manufacturer’s instructions. 2 4 2 4 and 4.2 mM NaHCO ). 25 μM CM-H DCFDA in phenol Absorbance (450 nm) was measured on a microplate reader 3 2 red-free DMEM was added to each sample and incubated (Thermo LabSystems, Multiskan Spectrum). Relative kinase for 25 minutes at 37 C, protected from light. To label nuclei, activity was calculated as follows: bisbenzimide (0.2 μg/mL, Thermo Fisher) was added to each sample for an additional 5 minutes. After thorough washing in HBSS buffer, samples were immersed in warmed phenol Relative kinase activity red-free DMEM and imaged in an Olympus Fluoview 1000 Average absorbance − Average absorbance sample blank microscope as confocal z-stacks (1 μm step size). = Quantity of crude protein used per assay 2.5. Confocal Microscopy and Image Analysis. Endothelial cells on PA gels were imaged by confocal microscopy and analyzed using MATLAB. In vitro cell samples were rinsed 2.7. Statistical Analysis. All statistical analyses were once with ice-cold PBS and fixed with ice-cold 4% parafor- conducted using MATLAB’s statistical toolbox. Graphs maldehyde. Samples were then permeabilized with 0.2% represent mean ± standard deviation. Multiple groups were Triton X-100 in PBS for 15 minutes and blocked with 1% compared using either two-way or n-way ANOVA with post bovine serum albumin (BSA) in PBS for 1 hour at room tem- hoc Tukey-Kramer test, and two groups were compared by perature. After fixation, mouse aortae were simultaneously Student’s t-test. Within each PKC assay, conditions were blocked and permeabilized in PBS containing 1% BSA and tested in duplicate. For ROS measurement, conditions were 0.3% Triton X-100. Endothelial cells on PA gels were labeled tested in triplicate. All experiments were conducted at least with primary antibodies for VE-cadherin (1 : 200, Santa two times, with at least three samples per condition. Cruz), β-catenin (1 : 200, Thermo Fisher), or pMLC (1 : 200, Cell Signaling) in 1% BSA in PBS overnight at 4 C. After sev- eral rinses with PBS, samples were then incubated with the 3. Results appropriate Alexa Fluor 488 or 633 (1 : 200, Thermo Fisher) secondary antibody, rhodamine phalloidin (16.5 nM, 3.1. Subendothelial Stiffness Was Higher in Eln+/- as Invitrogen), and bisbenzimide (0.2 μg/mL) for 1 hour at Compared to WT Mouse Aorta. Macroscale arterial stiffness, room temperature, protected from light. Samples were rinsed measured by pulse wave velocity or pressure myography, twice with PBS and mounted in 1 : 1 glycerol : PBS. Confocal increases in mice genetically engineered to produce less z-stacks were acquired for all samples with either a 0.25 or elastin (Eln+/-) [36, 37]. However, aortic stiffness had 0.5 μm step size (for in vitro and ex vivo samples, respec- not been characterized by atomic force microscopy in this tively) using an Olympus Fluoview 1000 confocal micro- mouse model. The longitudinally dissected mouse aorta scope at 60x magnification. (Figure 1(a)) shows the mounting technique and the thoracic A custom MATLAB code was created to quantify ROS and abdominal sections. Aortae with the intact endothelium and pMLC. The background was subtracted using a 50 × 50 were first labeled for β-catenin (green) to show endothelial pixel area. Images were then binarized using the same thresh- cell-cell junctions, actin (red) and nuclei (blue) to highlight old as determined using Otsu’s method, which calculates a cell structure, and collagen IV (white) to view the basement threshold based on pixel intensity distribution [35]. Noise membrane (Figure 1(b), top). When we removed the endo- was removed from binarized images by excluding small thelium from the longitudinally dissected mouse aortae, we objects (less than 9 pixels for ROS, less than 30 pixels for no longer observed β-catenin, confirming that cells were pMLC). The number of remaining pixels with intensities removed. The collagen IV layer remained intact and contig- above the threshold (“positive” pixels) was counted for uous, indicating that the subendothelial matrix remained each image. Three images per sample were quantified intact (Figure 1(b), bottom; collagen IV integrity was most using this method and averaged to quantify ROS or pMLC clear in the cross-sectional image). However, it is possible in each sample. that endothelial removal did damage the subendothelial layer, as evidenced by the spaces in the fluorescently labeled 2.6. PKC Activity Assay. After treatment, cells on PA gels samples. We therefore repeated the atomic force microscopy were quickly rinsed with ice-cold PBS and inverted onto in both scraped and unscraped WT mouse aortae and found 50 μL lysis buffer (20 mM MOPS, 50 mM β-glycerophos- no difference in aortic stiffness measurements. When aortic phate, 50 mM sodium fluoride, 1 mM sodium orthovanadate, samples were indented by AFM, the thoracic and abdominal 5 mM EGTA, 2 mM EDTA, 1% NP40, 1 mM dithiothreitol, aortic subendothelium from Eln+/- mice was about 1.75-fold 1 mM benzamidine, 1 mM phenylmethanesulfonyl fluoride, stiffer than that of WT mice (Figure 1(c)). The thoracic 10 μg/mL leupeptin, and 10 μg/mL aprotinin) for 10 minutes aorta was consistently stiffer than the abdominal aorta in at 4 C. Lysed cells were then scraped from the gel substrates, both genotypes. Applied Bionics and Biomechanics 5 Untreated PMA tBHP Nuclei ROS Nuclei ROS Nuclei ROS 2.0 1.5 1.0 0.5 0.0 6 kPa 14 kPa 29 kPa Substrate stiffness Figure 2: ROS increased with substrate stiffness following PMA treatment. PAEC monolayers on 6, 14, and 29 kPa gels were treated with 1 μM PMA for 10 minutes. Tert-butyl hydroperoxide (tBHP) was the positive control. Cell nuclei were labeled with Hoechst, and ROS with CM-H DCFDA. Samples were imaged at 20x by confocal microscopy. Scale bar is 25 μm. Oxidative stress was quantified using the number of positive pixels (above the threshold) using the custom MATLAB code. PMA-treated samples were normalized to untreated samples on the same substrate stiffness. The effect of substrate stiffness was significant by one-way ANOVA (p <0 01). p <0 05 and p <0 01 by post hoc Tukey-Kramer test. 3.2. PMA-Induced Increased Oxidative Stress in Endothelial significantly whether the cells were on soft or stiff PA gels Cells on the Stiffest PA Gels. We then created 6, 14, and (Figure 3(b)). Therefore, the PMA-induced differences in 29 kPa PA gels, which correspond to aortic stiffnesses in oxidative stress on stiffer substrates were not related to WT and Eln+/- mice, to determine if different ROS levels PKC activation. were produced by PMA-treated endothelial monolayers cultured on substrates of varied stiffness. Each sample was 3.4. Endothelial Cells Formed More Actin Stress Fibers in treated with 1 μM PMA for 10 minutes, based on preliminary Response to PMA on Stiffer Gels. ROS lead to endothelial experiments showing maximum cell viability and ROS at this actin fiber formation [40]. PMA-stimulated cells on increas- dose and time, and consistently imaged by confocal micros- ing stiffness substrates were labeled for actin fibers to copy. ROS were statistically similar following PMA treatment determine if substrate stiffness-dependent oxidative stress of cells on 6 and 14 kPa gels. In addition, endothelial cells on increased actin fiber formation. In untreated samples, actin stiff substrates did not show any baseline increase in ROS. fibers were primarily located around the cell periphery on However, in endothelial cells on 29 kPa substrates that were all substrates, although the effect was more pronounced in treated with PMA, ROS increased by more than 50% cells on the softest 6 kPa gels (Figure 4(a), representative cell (p <0 magnification in Figure 4(b)). Following 15 minutes of PMA 01 compared to fold change in cells on 6 kPa gels, Figure 2). Substrate stiffness effects on the PMA-induced fold treatment, actin fibers appeared in cells on the 14 and 29 kPa change in ROS were also significant by one-way ANOVA gels, but not in cells on the 6 kPa gels. This effect was even (p <0 01). These results demonstrate that stiffer substrates more pronounced following 30 minutes of PMA treatment, increase endothelial ROS in response to PMA. with larger stress fibers and nearly complete peripheral actin loss in endothelial cells on 14 and 29 kPa gels. Cells on 6 kPa 3.3. Endothelial Cell PKC Increased in Response to PMA gels largely retained peripheral actin with PMA treatment. Independent of PA Gel Stiffness. PMA induces ROS produc- PAECs were then labeled for pMLC to determine tion through PKC signaling [38, 39]. We therefore measured whether ROS-induced actin stress fiber formation was PKC activity in PMA-treated endothelial cells on 6, 14, and associated with increased actinomyosin contractility through pMLC localization to actin stress fibers. 1 μM PMA treat- 29 kPa PA gels to determine if PKC activation increased on stiffer substrates. PKC activity in PAEC increased 3-4-fold ment for 15 or 30 minutes did not induce pMLC transloca- within 5 minutes of PMA treatment (Figure 3(a)). However, tion to actin fibers or increase overall pMLC (Figure 5). In PKC activity in cells stimulated with PMA did not change contrast, the positive control (10 U/mL thrombin for 30 29 kPa 14 kPa 6 kPa Fold change in ROS (PMA/untreated) 6 Applied Bionics and Biomechanics 1.5 1.0 # ⁎ 휇 0.5 0 0.0 Untreated PMA 6 kPa 14 kPa 29 kPa (a) (b) Figure 3: PKC activity increased with PMA independent of substrate stiffness. (a) PKC activity was measured following 1 μM PMA treatment in endothelial cells on glass substrates. Purified active PKC was the positive control. p <0 01 compared to that untreated by Student’s t-test. (b) PAEC monolayers on 6, 14, and 29 kPa substrates were treated with 1 μM PMA for 5 minutes. PMA was significant by two-way ANOVA (p <0 001), but substrate stiffness was not significant. p <0 05 by post hoc Tukey-Kramer test compared to that untreated. Untreated 15 min PMA 30 min PMA Untreated 15 min PMA 30 min PMA (a) (b) Figure 4: Actin stress fiber formation was greater in endothelial cells on stiffer substrates following PMA. PAEC monolayers on 6, 14, or 29 kPa gels were treated with 1 μM PMA for 15 or 30 minutes prior to fixation and immunofluorescent labeling of actin (rhodamine phalloidin). (a) Maximum intensity projection from confocal z-stacks at 60x magnification. Scale bar is 25 μm. (b) Magnified representative cells from (a). minutes) increased overall pMLC approximately 14-fold, 3.5. Adherens Junctions Became Less Reticular in Response to with pMLC localized along the actin fibers. These results PMA on Stiffer Gels. ROS induce cell-cell junction loss, which indicate that PMA-induced actin fiber formation in cells has been attributed in part to adherens junction protein phos- on stiffer substrate did not correspond to actinomyosin phorylation and internalization [41–43]. We therefore mea- contractility. sured if stiff substrates exacerbate ROS-mediated endothelial 29 kPa 14 kPa 6 kPa Absorbance (a.u.) Blank Active PKC Untreated PMA, 5 min PMA, 10 min 29 kPa 14 kPa 6 kPa Normalized PKC activity Absorbance (a.u./g protein) Applied Bionics and Biomechanics 7 Untreated 15 min PMA 30 min PMA 6 kPa 14 kPa 29 kPa Substrate stiffness Untreated 15 min PMA 30 min PMA Thrombin Thrombin (positive control) Figure 5: PMA treatment did not increase pMLC localization to actin stress fibers in cells on varied stiffness substrates. PAEC monolayers on 6, 14, or 29 kPa gels were treated with 1 μM PMA for 15 or 30 minutes, prior to fixation and pMLC immunofluorescent labeling. For the positive control, cells on a 29 kPa gel were treated with 10 U/mL thrombin for 30 minutes. Images are maximum intensity projections from 60x confocal z-stacks. Scale bar is 25 μm. pMLC-positive pixels were quantified using the custom MATLAB code. Stiffness and PMA treatment were not significant by n-way ANOVA. Untreated 15 min PMA 30 min PMA Untreated 15 min PMA 30 min PMA (a) (b) Figure 6: Reticular adherens junction loss was greater in cells on stiffer substrates following PMA. PAEC monolayers on 6, 14, or 29 kPa gels were treated with 1 μM PMA for 15 or 30 minutes, prior to fixation and immunofluorescent labeling of the cell-cell junction protein β-catenin. (a) Maximum intensity projection from confocal z-stacks at 60x magnification. Scale bar is 25 μm. (b) Representative junctions. adherens junction loss in response to PMA. In untreated representative magnified junctions in Figure 6(b)). In con- cells on 6, 14, and 29 kPa gels, wide reticular adherens junc- trast to what has been observed in other published work, tions were evident between adjacent cells (Figure 6(a), we did not observe any changes in adherens junction based 29 kPa 14 kPa 6 kPa 29 kPa 14 kPa 6 kPa 29 kPa 14 kPa 6 kPa pMLC-positive (thousands) 8 Applied Bionics and Biomechanics 훽-Catenin Untreated PMA (a) (b) 훽-Catenin Actin Untreated PMA Untreated PMA (c) (d) Figure 7: ROS scavengers prevented PMA-induced adherens junction loss and actin fiber redistribution. PAEC monolayers were pretreated with ROS scavengers (4 mM N-acetyl cysteine, 50 mM sodium pyruvate) for 1 hour before the 30-minute treatment with 1 μM PMA. Samples were fixed and immunofluorescently labeled for (a) β-catenin, with representative cells magnified in (c), and immunofluorescently labeled for (b) actin, with representative cells magnified in (d). Images are maximum intensity projections from 60x confocal z-stacks. Scale bar is 25 μm. on substrate stiffness alone perhaps due to the use of a differ- junction morphology change was abrogated. More strikingly, ent endothelial cell type [12]. After 15 or 30 minutes of PMA ROS scavengers prevented PMA-induced actin reorgani- treatment, reticular junctions were mostly maintained in cells zation (Figure 7(b), representative cell magnification in on the 6 kPa gels. In contrast, cells on the stiffest 29 kPa sub- Figure 7(d)). Cells treated with the ROS scavengers prior strates lost most junction reticular structures and instead had to PMA showed peripheral actin bands which were similar to linear or disrupted cell-cell junctions. These results demon- those in untreated cells. Thus, oxidative stress was likely responsible for PMA-induced reticular junction loss and strate that endothelial reticular junction structure loss is exacerbated by stiffer substrates in response to the ROS- actin fiber formation. stimulant PMA. To confirm that ROS were responsible for the PMA- 4. Discussion induced changes in adherens junctions and actin fiber formation, the ROS scavengers N-acetyl cysteine and sodium Oxidative stress and more specifically the enzyme responsi- pyruvate were administered for 1 hour prior to PMA treat- ble for superoxide production, NADPH oxidase, have been ment. These experiments were performed on glass coverslips, implicated in cardiovascular disease and atherosclerosis in since junction loss following PMA exposure was highest on particular [44, 45]. We now show that stiffer substrates exac- these stiffest substrates. ROS scavengers themselves did not erbate endothelial cell oxidative stress. In response to PMA, affect cell-cell junction structure, and cells treated with endothelial cells on the stiffest substrates showed more ROS PMA alone showed linear and invaginated adherens junctions and actin stress fibers and showed greater adherens junction (Figure 7(a), representative cell magnification in Figure 7(c)). loss, which was not attributed to cell contractility. Stiffer In PAEC pretreated with ROS scavengers prior to PMA, the aortas from Eln+/- mice also showed less VE-cadherin at ROS Scavenger None ROS Scavenger None ROS Scavenger None Applied Bionics and Biomechanics 9 stiffness. However, while these morphological responses cell-cell membranes and increased peripheral actin fiber formation in response to PMA. Since PMA-induced PKC to substrate stiffness are no longer observed as endothelial activation was not affected by substrate stiffness, it is likely cells reach confluency, this study shows that both endothe- that substrate stiffness affected cells through alternative lial biochemical responses and cell-cell interactions do pathways. These data suggest that oxidative stress and its det- change with substrate stiffness. rimental downstream effects on endothelial cells and vascular ROS, specifically superoxide and its byproduct hydrogen disease may be enhanced in stiffer arteries. peroxide, have been shown to regulate actin fibers in vascular The vascular mechanics of the elastin haploinsufficient cells [40, 72–74]. Actin fiber formation in subconfluent mouse have been extensively studied, both in terms of passive reoxygenated hypoxic aortic endothelial cells was inhibited mechanical stretch in response to increasing pressure and in by superoxide dismutase overexpression, suggesting a key terms of vasodilation and constriction in response to role for superoxide [75]. Superoxide can reversibly oxidize biochemical stimuli [13, 37, 46, 47]. These studies focused proteins, including actin itself. In endothelial cells, actin primarily on the decrease in total elastin in the vascular wall, oxidation may be essential for actin polymerization during as well as the increase in elastin lamellae. We and others did cell migration. For example, in migrating mouse aortic endo- not find any changes in other extracellular matrix proteins, in thelial cells, actin monomer incorporation into actin fibers particular collagen, which is the other primary protein was diminished following treatment with the NADPH oxi- thought to define vascular wall stiffness. However, some dase inhibitor diphenyleneiodonium (DPI) and a superoxide recent studies in other tissues in elastin haploinsufficient mice dismutase mimetic [76]. Alternatively, superoxide can oxi- suggest that there are also collagen changes in these animals. dize RhoA, enhancing GDP dissociation and subsequent The lungs of Eln +/- mice contained nearly twice as much col- activation. Fibroblasts with an oxidation-resistant form of lagen 1 and lysyl oxidase, an important collagen crosslinker, as RhoA did not form stress fibers in response to hydrogen WT mice [48]. Achilles tendons of Eln+/- mice had the same peroxide [77]. While the source of increased ROS in endo- total collagen content as WT mice but different collagen fibril thelial cells on stiffer substrates remains unknown, we diameter distribution [49]. Thus, it is possible that the hypothesize that stiff substrates increase NADPH oxidase increased stiffness we measured in the Eln+/- aorta relates production or assembly, since NADPH oxidase appears to changes in collagen content and/or structure. to produce the most nonmitochondrial superoxide in The thoracic aorta was consistently stiffer than the endothelial cells [78]. We hope to investigate this mechanism abdominal aorta in both WT and Eln+/- mice. These data further in future studies. agree with human studies in which aortic stiffness decreased Although endothelial oxidative stress has not been with distance from the heart, especially in older patients studied on varied stiffness substrates, endothelial superoxide [50, 51]. Other studies in C57BL/6 mice demonstrated that production is mechanosensitive, specifically in response to the aortic elastic modulus was highest in the distal thoracic shear stress [79]. Bovine aortic endothelial cells produced aorta and lowest in the abdominal aorta [52]. In a subsequent three times more superoxide under oscillatory shear stress study, it was shown that the decrease in aortic stiffness along compared to laminar flow [80]. Shear stress activates Rac, the length of the aorta was accounted for by a decrease in which is downstream of integrin activation and contributes total and lamellar elastin [53]. Since our data show a propor- to ROS production [81, 82]. Epithelial cells have been shown tionally similar decrease in aortic stiffness from the thoracic to produce more ROS when on stiffer substrates. MMP-3- to the abdominal sections in both WT and Eln+/- mice, stimulated ROS production was approximately 3.5-fold higher in epithelial cells on 4.02 kPa substrates compared to it is likely that elastin content is important to the regional stiffness variation. 0.13 kPasubstrates;β1integrinsubunitknockdowneliminated Our data support other recently published studies show- ROSproductioninresponsetoMMP-3[83].These findings sug- ing that substrate stiffness affects not only single endothelial gest that integrin activation-induced Rac1 activity contrib- cells but also confluent endothelial monolayers [12]. In vitro utes to ROS production in cells on stiffer substrates [84]. studies of cell response to substrate stiffness began when The increase in adherens junction disruption could be Pelham and Wang first used protein-coated PA gels to show either contractility-dependent or contractility-independent. that both rat kidney epithelial and 3T3 fibroblasts spread Permeability agents, including thrombin, lipopolysaccha- to a greater extent on stiff than soft substrates [54]. Since ride (LPS), TNF-α, and vascular endothelial growth factor that seminal paper, many cell types were shown to change (VEGF), activate the Rho/ROCK pathway and cell con- their morphology [55–58], motility [59, 60], differentiation tractility [85]. The ROCK inhibitor Y-27632 prevented [61, 62], and proliferation [63, 64] in response to substrate adherens junction disruption in endothelial monolayer stiffness. For endothelial cells specifically, single cells increase studies, although some effects may be endothelial the spread area [65, 66], stiffness [67], cell-matrix and cell- subtype-dependent (e.g., macrovascular or microvascular) cell forces [66, 68], and proliferation [69] with substrate stiff- [86, 87]. In epithelial cell protrusions, H O increased 2 2 ness. However, as cells proliferated and reached confluency, actin polymerization, cofilin activity, and barbed ends; substrate-dependent differences were diminished or no however, myosin IIA did not colocalize with actin fibers in longer observed [70, 71]. Similarly, we did not observe H O -treated cell protrusions [88]. These data fit with our 2 2 any changes in endothelial cell and actin stress fiber mor- results that actin contractility did not increase with oxidative phology, focal adhesion size, or focal adhesion number in stress. Therefore, it is more likely in our case that oxidative endothelial cell monolayers on substrates of different stress induced contractility-independent adherens junction 10 Applied Bionics and Biomechanics disruption. ROS also disrupt cell-cell junctions through 5. Conclusions VE-cadherin phosphorylation. Endothelial cell treatment This work illustrates a novel potential mechanism for with permeability agonists, such as VEGF and histamine, substrate-enhanced oxidative stress in response to PKC resulted in VE-cadherin tyrosine phosphorylation [89, 90]. activation in the endothelium. Since many endothelial cell In HUVECs, the ROS scavenger N-acetyl cysteine prevented studies are performed on tissue culture polystyrene of VE-cadherin phosphorylation by TNF-α [91]. Thus, we essentially infinite stiffness, these studies may overestimate hypothesize that adherens junction protein phosphorylation endothelial cell response to stressors. Further study of the resulted in cell-cell junction loss, although we did not directly interaction between arterial stiffness and oxidative stress measure it. could improve therapies to prevent or ameliorate endothelial ROS can also lead to adverse effects on the endothelium barrier dysfunction. beyond adherens junction loss. For example, superoxide (O ) interacts with nitric oxide (NO) to form peroxynitrite. Data Availability This interaction effectively decreases the NO availability, which is needed for vasodilation [92, 93]. Superoxide also AFM indentation curves, PKC quantification, and immuno- uncouples eNOS by oxidizing the eNOS cofactor tetrahydro- fluorescence images used to support the finding of this study biopterin (BH )[94–96]. Uncoupled eNOS produces more may be released upon application to the corresponding superoxide instead of NO [97], which further increases author. peroxynitrite. Protein nitration by peroxynitrite is widely observed in many cardiovascular diseases [98]. Thus, vascu- Conflicts of Interest lar stiffness-induced endothelial oxidative stress could have damaging effects beyond endothelial barrier function. The authors declare that there is no conflict of interest Substrate stiffness also affects other cell types beyond regarding the publication of this paper. endothelial cells, including fibroblasts, breast cancer cells, and stem cells [99, 100]. In vivo, tumors are stiffer than their Authors’ Contributions surrounding environment, which may alter both basal func- tion and inflammatory response in all of these cell types. In RLU helped design the study, carried out all experiments, addition, some tumors overexpress specific NADPH oxidases helped analyze the data, and drafted the manuscript. SS [101]. This overexpression could couple with increased tumor assisted with data analysis. AMC helped design the study, stiffness to further magnify oxidative stress in tumors. Tumor analyzed the data, and drafted the manuscript. All authors oxidative stress contributes to tissue injury and DNA damage gave final approval for publication. in premalignant conditions, as well as in cancer initiation and progression. Since the tumor cells themselves may be resistant Acknowledgments to oxidative stress, the injury to the surrounding tissue may be more severe [102]. Thus, stiffness-associated ROS inhibition This work was supported by the American Heart Association could potentially decrease cancer-induced damage and (grant number SDG4460068). We thank Patrick Osei-Owusu inhibit cancer metastasis through compromised blood vessels. for providing the WT and Eln+/- mouse aortae. While our work shows that PMA-induced oxidative stress increases with substrate stiffness, our research is not References without limitations. Phorbol esters, including PMA, are found in certain plants and can cause toxicity in animals [1] S. Laurent, J. Cockcroft, L. van Bortel et al., “Expert consensus when consumed [38]. Yet, PMA itself is not involved in document on arterial stiffness: methodological issues and clinical applications,” European Heart Journal, vol. 27, cardiovascular disease pathogenesis. We chose to use PMA no. 21, pp. 2588–2605, 2006. to isolate ROS production from cell contractility in cell-cell [2] C. Vlachopoulos, K. Aznaouridis, and C. Stefanadis, “Predic- junction loss [25, 26]; however, these studies should be tion of cardiovascular events and all-cause mortality with repeated using a physiologically relevant ROS inducer arterial stiffness: a systematic review and meta-analysis,” (e.g., tumor necrosis factor-α). PA gels do not fully capture Journal of the American College of Cardiology, vol. 55, the endothelial mechanical environment, including substrate no. 13, pp. 1318–1327, 2010. viscoelasticity and relaxation as well as shear stress or strain. [3] Y. Ben-Shlomo, M. Spears, C. Boustred et al., “Aortic pulse Several recent papers have shown interactions between shear wave velocity improves cardiovascular event prediction: an stress and substrate stiffness, demonstrating that softer sub- individual participant meta-analysis of prospective observa- strates augmented the atheroprotective effects of laminar tional data from 17,635 subjects,” Journal of the American shear stress [103, 104]. We also were not able to specifically College of Cardiology, vol. 63, no. 7, pp. 636–646, 2014. detect superoxide in intact vessels under physiological [4] A. J. Bank, H. Wang, J. E. Holte, K. Mullen, R. Shammas, and conditions due to technical challenges. In future studies, S. H. Kubo, “Contribution of collagen, elastin, and smooth a superoxide-specific indicator such as dihydroethidium muscle to in vivo human brachial artery wall stress and elastic could be used along with NADPH oxidase component knock- modulus,” Circulation, vol. 94, no. 12, pp. 3263–3270, 1996. down to support the hypothesis that stiffness-induced integ- [5] B. S. Fleenor, “Large elastic artery stiffness with aging: novel rin activation enhances superoxide production by NADPH translational mechanisms and interventions,” Aging and oxidase [105]. Disease, vol. 4, no. 2, pp. 76–83, 2013. Applied Bionics and Biomechanics 11 [21] F. Cosentino, M. Eto, P. de Paolis et al., “High glucose causes [6] M. Veerasamy, G. A. Ford, D. Neely et al., “Association of aging, arterial stiffness, and cardiovascular disease: a review,” upregulation of cyclooxygenase-2 and alters prostanoid Cardiology in Review, vol. 22, no. 5, pp. 223–232, 2014. profile in human endothelial cells: role of protein kinase C and reactive oxygen species,” Circulation, vol. 107, no. 7, [7] C. Xu, C. K. Zarins, P. S. Pannaraj, H. S. Bassiouny, and pp. 1017–1023, 2003. S. Glagov, “Hypercholesterolemia superimposed by experi- mental hypertension induces differential distribution of [22] T. Inoguchi, P. Li, F. Umeda et al., “High glucose level and collagen and elastin,” Arteriosclerosis, Thrombosis, and free fatty acid stimulate reactive oxygen species production Vascular Biology, vol. 20, no. 12, pp. 2566–2572, 2000. through protein kinase C–dependent activation of NAD(P)H oxidase in cultured vascular cells,” Diabetes, vol. 49, no. 11, [8] M. Cecelja and P. Chowienczyk, “Dissociation of aortic pulse pp. 1939–1945, 2000. wave velocity with risk factors for cardiovascular disease other than hypertension: a systematic review,” Hypertension, [23] L. Quagliaro, L. Piconi, R. Assaloni, L. Martinelli, E. Motz, vol. 54, no. 6, pp. 1328–1336, 2009. and A. Ceriello, “Intermittent high glucose enhances apoptosis related to oxidative stress in human umbilical vein [9] A. Scuteri, P. G. Cunha, E. A. Rosei et al., “Arterial stiffness endothelial cells: the role of protein kinase C and NAD(P)H- and influences of the metabolic syndrome: a cross-countries oxidase activation,” Diabetes, vol. 52, no. 11, pp. 2795–2804, study,” Atherosclerosis, vol. 233, no. 2, pp. 654–660, 2014. [10] G. F. Mitchell, “Arterial stiffness: insights from Framingham [24] M. T. Elnakish, H. H. Hassanain, P. M. Janssen, M. G. and Iceland,” Current Opinion in Nephrology and Hyperten- Angelos, and M. Khan, “Emerging role of oxidative stress sion, vol. 24, no. 1, pp. 1–7, 2015. in metabolic syndrome and cardiovascular diseases: impor- [11] B. M. Kaess, J. Rong, M. G. Larson et al., “Aortic stiffness, tant role of Rac/NADPH oxidase,” The Journal of Pathol- blood pressure progression, and incident hypertension,” ogy, vol. 231, no. 3, pp. 290–300, 2013. Journal of the American Medical Association, vol. 308, no. 9, [25] J. G. N. Garcia, H. W. Davis, and C. E. Patterson, “Regulation pp. 875–881, 2012. of endothelial cell gap formation and barrier dysfunction: [12] J. Huynh, N. Nishimura, K. Rana et al., “Age-related intimal role of myosin light chain phosphorylation,” Journal of Cellu- stiffening enhances endothelial permeability and leukocyte lar Physiology, vol. 163, no. 3, pp. 510–522, 1995. transmigration,” Science Translational Medicine, vol. 3, [26] N. V. Bogatcheva, A. D. Verin, P. Wang et al., “Phorbol esters no. 112, article 112ra122, 2011. increase MLC phosphorylation and actin remodeling in [13] P. Osei-Owusu, R. H. Knutsen, B. A. Kozel, H. H. Dietrich, bovine lung endothelium without increased contraction,” K. J. Blumer, and R. P. Mecham, “Altered reactivity of resis- American Journal of Physiology-Lung Cellular and Molecular tance vasculature contributes to hypertension in elastin insuf- Physiology, vol. 285, no. 2, pp. L415–L426, 2003. ficiency,” American Journal of Physiology-Heart and [27] A. B. Moy, K. Blackwell, N. Wang et al., “Phorbol Circulatory Physiology, vol. 306, no. 5, pp. H654–H666, 2014. ester-mediated pulmonary artery endothelial barrier dys- [14] A. A. Birukova, X. Tian, I. Cokic, Y. Beckham, M. L. Gardel, function through regulation of actin cytoskeletal mechanics,” and K. G. Birukov, “Endothelial barrier disruption and recov- American Journal of Physiology-Lung Cellular and Molecular ery is controlled by substrate stiffness,” Microvascular Physiology, vol. 287, no. 1, pp. L153–L167, 2004. Research, vol. 87, pp. 50–57, 2013. [28] J. E. Stasek, C. E. Patterson, and J. G. N. Garcia, “Protein- [15] R. Krishnan, D. D. Klumpers, C. Y. Park et al., “Substrate kinase-c phosphorylates caldesmon77 and vimentin and stiffening promotes endothelial monolayer disruption enhances albumin permeability across cultured bovine through enhanced physical forces,” American Journal of pulmonary-artery endothelial-cell monolayers,” Journal of Physiology-Cell Physiology, vol. 300, no. 1, pp. C146–C154, Cellular Physiology, vol. 153, no. 1, pp. 62–75, 1992. [29] N. V. Bogatcheva, A. Birukova, T. Borbiev et al., “Caldesmon [16] G. Faury, M. Pezet, R. H. Knutsen et al., “Developmental is a cytoskeletal target for PKC in endothelium,” Journal of adaptation of the mouse cardiovascular system to elastin Cellular Biochemistry, vol. 99, no. 6, pp. 1593–1605, 2006. haploinsufficiency,” Journal of Clinical Investigation, vol. 112, no. 9, pp. 1419–1428, 2003. [30] A. J. Oliveira-dos-Santos, G. Matsumoto, B. E. Snow et al., “Regulation of T cell activation, anxiety, and male aggression [17] C. M. McEniery, S. Wallace, I. S. Mackenzie et al., “Endothe- by RGS2,” Proceedings of the National Academy of Sciences of lial function is associated with pulse pressure, pulse wave the United States of America, vol. 97, no. 22, pp. 12272– velocity, and augmentation index in healthy humans,” Hyper- 12277, 2000. tension, vol. 48, no. 4, pp. 602–608, 2006. [31] J. Peloquin, J. Huynh, R. M. Williams, and C. A. Reinhart- [18] A. Nigam, G. F. Mitchell, J. Lambert, and J. C. Tardif, “Rela- King, “Indentation measurements of the subendothelial tion between conduit vessel stiffness (assessed by tonometry) matrix in bovine carotid arteries,” Journal of Biomechanics, and endothelial function (assessed by flow-mediated dilata- vol. 44, no. 5, pp. 815–821, 2011. tion) in patients with and without coronary heart disease,” The American Journal of Cardiology, vol. 92, no. 4, pp. 395– [32] J. Solon, I. Levental, K. Sengupta, P. C. Georges, and P. A. Janmey, “Fibroblast adaptation and stiffness matching to 399, 2003. soft elastic substrates,” Biophysical Journal, vol. 93, no. 12, [19] A. M. Briones and R. M. Touyz, “Oxidative stress and hyper- pp. 4453–4461, 2007. tension: current concepts,” Current Hypertension Reports, vol. 12, no. 2, pp. 135–142, 2010. [33] J. R. Tse and A. J. Engler, “Preparation of hydrogel substrates with tunable mechanical properties,” Current Protocols in [20] P. D. Ray, B.-W. Huang, and Y. Tsuji, “Reactive oxygen Cell Biology, vol. 47, no. 1, pp. 10.16.1–10.16.16, 2010. species (ROS) homeostasis and redox regulation in cellular signaling,” Cellular Signalling, vol. 24, no. 5, pp. 981–990, [34] R. L. Urbano, C. Furia, S. Basehore, and A. M. Clyne, “Stiff 2012. substrates increase inflammation-induced endothelial 12 Applied Bionics and Biomechanics monolayer tension and permeability,” Biophysical Journal, [49] J. D. Eekhoff, F. Fang, L. G. Kahan et al., “Functionally dis- vol. 113, no. 3, pp. 645–655, 2017. tinct tendons from elastin haploinsufficient mice exhibit mild stiffening and tendon-specific structural alteration,” Journal [35] N. Otsu, “A threshold selection method from gray-level of Biomechanical Engineering, vol. 139, no. 11, p. 111003, histograms,” IEEE Transactions on Systems, Man, and Cyber- netics, vol. 9, no. 1, pp. 62–66, 1979. [50] V. Taviani, S. S. Hickson, C. J. Hardy et al., “Age-related [36] J. E. Wagenseil, C. H. Ciliberto, R. H. Knutsen, M. A. Levy, changes of regional pulse wave velocity in the descending A. Kovacs, and R. P. Mecham, “Reduced vessel elasticity aorta using Fourier velocity encoded M-mode,” Magnetic alters cardiovascular structure and function in newborn Resonance in Medicine, vol. 65, no. 1, pp. 261–268, 2011. mice,” Circulation Research, vol. 104, no. 10, pp. 1217–1224, 2009. [51] W. J. Rogers, Y. L. Hu, D. Coast et al., “Age-associated changes in regional aortic pulse wave velocity,” Journal of [37] L. Carta, J. E. Wagenseil, R. H. Knutsen et al., “Discrete con- the American College of Cardiology, vol. 38, no. 4, pp. 1123– tributions of elastic fiber components to arterial development 1129, 2001. and mechanical compliance,” Arteriosclerosis, Thrombosis, [52] X. Guo and G. S. Kassab, “Variation of mechanical properties and Vascular Biology, vol. 29, no. 12, pp. 2083–2089, 2009. along the length of the aorta in C57bl/6 mice,” American [38] G. Goel, H. P. S. Makkar, G. Francis, and K. Becker, “Phorbol Journal of Physiology-Heart and Circulatory Physiology, esters: structure, biological activity, and toxicity in animals,” vol. 285, no. 6, pp. H2614–H2622, 2003. International Journal of Toxicology, vol. 26, no. 4, pp. 279– [53] T. E. Clark, M. A. Lillie, A. W. Vogl, J. M. Gosline, and R. E. 288, 2007. Shadwick, “Mechanical contribution of lamellar and interla- [39] J. M. Li, A. M. Mullen, S. Yun et al., “Essential role of the phox mellar elastin along the mouse aorta,” Journal of Biomechan- NADPH oxidase subunit p47 in endothelial cell superox- ics, vol. 48, no. 13, pp. 3599–3605, 2015. ide production in response to phorbol ester and tumor necro- [54] R. J. Pelham and Y.-L. Wang, “Cell locomotion and focal sis factor-α,” Circulation Research, vol. 90, no. 2, pp. 143–150, adhesions are regulated by substrate flexibility,” Proceedings of the National Academy of Sciences of the United States of [40] H. Lum and K. A. Roebuck, “Oxidant stress and endothelial America, vol. 94, no. 25, pp. 13661–13665, 1997. cell dysfunction,” American Journal of Physiology-Cell [55] J. P. Califano and C. A. Reinhart-King, “A balance of sub- Physiology, vol. 280, no. 4, pp. C719–C741, 2001. strate mechanics and matrix chemistry regulates endothelial [41] S. van Wetering, J. D. van Buul, S. Quik et al., “Reactive cell network assembly,” Cellular and Molecular Bioengineer- oxygen species mediate Rac-induced loss of cell-cell adhesion ing, vol. 1, no. 2-3, pp. 122–132, 2008. in primary human endothelial cells,” Journal of Cell Science, [56] A. Engler, L. Bacakova, C. Newman, A. Hategan, M. Griffin, vol. 115, no. 9, pp. 1837–1846, 2002. and D. Discher, “Substrate compliance versus ligand density [42] P. V. Usatyuk and V. Natarajan, “Regulation of reactive oxy- in cell on gel responses,” Biophysical Journal, vol. 86, no. 1, gen species-induced endothelial cell-cell and cell-matrix con- pp. 617–628, 2004. tacts by focal adhesion kinase and adherens junction [57] A. J. Engler, C. Carag-Krieger, C. P. Johnson et al., “Embry- proteins,” American Journal of Physiology-Lung Cellular and onic cardiomyocytes beat best on a matrix with heart-like Molecular Physiology, vol. 289, no. 6, pp. L999–L1010, elasticity: scar-like rigidity inhibits beating,” Journal of Cell Science, vol. 121, no. 22, pp. 3794–3802, 2008. [43] C. G. Kevil, N. Ohno, D. C. Gute et al., “Role of cadherin [58] P. Moshayedi, L. da F Costa, A. Christ et al., “Mechanosensi- internalization in hydrogen peroxide-mediated endothelial tivity of astrocytes on optimized polyacrylamide gels ana- permeability,” Free Radical Biology and Medicine, vol. 24, lyzed by quantitative morphometry,” Journal of Physics: no. 6, pp. 1015–1022, 1998. Condensed Matter, vol. 22, no. 19, p. 194114, 2010. [44] G. R. Drummond, S. Selemidis, K. K. Griendling, and C. G. [59] P. W. Oakes, D. C. Patel, N. A. Morin et al., “Neutrophil mor- Sobey, “Combating oxidative stress in vascular disease: phology and migration are affected by substrate elasticity,” NADPH oxidases as therapeutic targets,” Nature Reviews Blood, vol. 114, no. 7, pp. 1387–1395, 2009. Drug Discovery, vol. 10, no. 6, pp. 453–471, 2011. [60] S. R. Peyton and A. J. Putnam, “Extracellular matrix rigidity [45] N. R. Madamanchi, A. Vendrov, and M. S. Runge, “Oxidative governs smooth muscle cell motility in a biphasic fashion,” stress and vascular disease,” Arteriosclerosis Thrombosis and Journal of Cellular Physiology, vol. 204, no. 1, pp. 198–209, Vascular Biology, vol. 25, no. 1, pp. 29–38, 2005. [46] J. E. Wagenseil, N. L. Nerurkar, R. H. Knutsen, R. J. Okamoto, [61] A. J. Engler, M. A. Griffin, S. Sen, C. G. Bönnemann, H. L. D. Y. Li, and R. P. Mecham, “Effects of elastin haploinsuffi- Sweeney, and D. E. Discher, “Myotubes differentiate opti- ciency on the mechanical behavior of mouse arteries,” Amer- mally on substrates with tissue-like stiffness,” The Journal of ican Journal of Physiology-Heart and Circulatory Physiology, Cell Biology, vol. 166, no. 6, pp. 877–887, 2004. vol. 289, no. 3, pp. H1209–H1217, 2005. [62] A. J. Engler, S. Sen, H. L. Sweeney, and D. E. Discher, “Matrix [47] B. A. Sather, D. Hageman, and J. E. Wagenseil, “Murray’s law elasticity directs stem cell lineage specification,” Cell, vol. 126, in elastin haploinsufficient (Eln+/-) and wild-type (WT) no. 4, pp. 677–689, 2006. mice,” Journal of Biomechanical Engineering, vol. 134, no. 12, p. 124504, 2012. [63] T. A. Ulrich, E. M. de Juan Pardo, and S. Kumar, “The mechanical rigidity of the extracellular matrix regulates the [48] A. Hilgendorff, K. Parai, R. Ertsey et al., “Lung matrix and structure, motility, and proliferation of glioma cells,” Cancer vascular remodeling in mechanically ventilated elastin Research, vol. 69, no. 10, pp. 4167–4174, 2009. haploinsufficient newborn mice,” American Journal of Physiology-Lung Cellular and Molecular Physiology, vol. 308, [64] S. R. Peyton, C. B. Raub, V. P. Keschrumrus, and A. J. no. 5, pp. L464–L478, 2015. Putnam, “The use of poly(ethylene glycol) hydrogels to Applied Bionics and Biomechanics 13 [79] R. P. Brandes, N. Weissmann, and K. Schröder, “Nox family investigate the impact of ECM chemistry and mechanics on smooth muscle cells,” Biomaterials, vol. 27, no. 28, NADPH oxidases in mechano-transduction: mechanisms pp. 4881–4893, 2006. and consequences,” Antioxidants & Redox Signaling, vol. 20, no. 6, pp. 887–898, 2014. [65] J. A. Wood, N. M. Shah, C. T. McKee et al., “The role of sub- stratum compliance of hydrogels on vascular endothelial cell [80] J. S. McNally, M. E. Davis, D. P. Giddens et al., “Role of behavior,” Biomaterials, vol. 32, no. 22, pp. 5056–5064, 2011. xanthine oxidoreductase and NAD(P)H oxidase in endothe- lial superoxide production in response to oscillatory shear [66] J. P. Califano and C. A. Reinhart-King, “Substrate stiffness stress,” American Journal of Physiology-Heart and Circula- and cell area predict cellular traction stresses in single cells tory Physiology, vol. 285, no. 6, pp. H2290–H2297, 2003. and cells in contact,” Cellular and Molecular Bioengineering, vol. 3, no. 1, pp. 68–75, 2010. [81] E. Tzima, M. del Pozo, W. B. Kiosses et al., “Activation of Rac1 by shear stress in endothelial cells mediates both cyto- [67] F. J. Byfield, R. K. Reen, T. P. Shentu, I. Levitan, and K. J. skeletal reorganization and effects on gene expression,” The Gooch, “Endothelial actin and cell stiffness is modulated by EMBO Journal, vol. 21, no. 24, pp. 6791–6800, 2002. substrate stiffness in 2D and 3D,” Journal of Biomechanics, vol. 42, no. 8, pp. 1114–1119, 2009. [82] L. H. Yeh, Y. J. Park, R. J. Hansalia et al., “Shear-induced tyrosine phosphorylation in endothelial cells requires Rac1- [68] V. Maruthamuthu, B. Sabass, U. S. Schwarz, and M. L. dependent production of ROS,” American Journal of Gardel, “Cell-ECM traction force modulates endogenous Physiology-Cell Physiology, vol. 276, no. 4, pp. C838–C847, tension at cell-cell contacts,” Proceedings of the National Academy of Sciences of the United States of America, vol. 108, no. 12, pp. 4708–4713, 2011. [83] K. Lee, Q. K. Chen, C. Lui, M. A. Cichon, D. C. Radisky, and C. M. Nelson, “Matrix compliance regulates Rac1b [69] Y. T. Yeh, S. S. Hur, J. Chang et al., “Matrix stiffness regulates localization, NADPH oxidase assembly, and epithelial- endothelial cell proliferation through septin 9,” PLoS One, mesenchymal transition,” Molecular Biology of the Cell, vol. 7, no. 10, 2012. vol. 23, no. 20, pp. 4097–4108, 2012. [70] T. Yeung, P. C. Georges, L. A. Flanagan et al., “Effects of [84] M. A. del Pozo, N. B. Alderson, W. B. Kiosses, H. H. Chiang, substrate stiffness on cell morphology, cytoskeletal structure, R. G. Anderson, and M. A. Schwartz, “Integrins regulate Rac and adhesion,” Cell Motility and the Cytoskeleton, vol. 60, targeting by internalization of membrane domains,” Science, no. 1, pp. 24–34, 2005. vol. 303, no. 5659, pp. 839–842, 2004. [71] K. M. Stroka and H. Aranda-Espinoza, “Endothelial cell [85] B. Wojciak-Stothard and A. J. Ridley, “Rho GTPases and the substrate stiffness influences neutrophil transmigration via regulation of endothelial permeability,” Vascular Pharmacol- myosin light chain kinase-dependent cell contraction,” Blood, ogy, vol. 39, no. 4-5, pp. 187–199, 2002. vol. 118, no. 6, pp. 1632–1640, 2011. [86] J. M. Carbajal, M. L. Gratrix, C. H. Yu, and R. C. Schaeffer Jr., [72] I. Dalle-Donne, R. Rossi, A. Milzani, P. di Simplicio, and “ROCK mediates thrombin’s endothelial barrier dysfunc- R. Colombo, “The actin cytoskeleton response to oxidants: tion,” American Journal of Physiology-Cell Physiology, from small heat shock protein phosphorylation to changes vol. 279, no. 1, pp. C195–C204, 2000. in the redox state of actin itself,” Free Radical Biology and Medicine, vol. 31, no. 12, pp. 1624–1632, 2001. [87] A. A. Birukova, K. Smurova, K. G. Birukov, K. Kaibuchi, J. G. N. Garcia, and A. D. Verin, “Role of Rho GTPases in [73] Q. Xu, L. P. Huff, M. Fujii, and K. K. Griendling, “Redox thrombin-induced lung vascular endothelial cells barrier dys- regulation of the actin cytoskeleton and its role in the vascular function,” Microvascular Research, vol. 67, no. 1, pp. 64–77, system,” Free Radical Biology and Medicine, vol. 109, pp. 84–107, 2017. [88] N. Taulet, V. D. Delorme-Walker, and C. DerMardirossian, [74] K. Buyukhatipoglu and A. M. Clyne, “Superparamagnetic “Reactive oxygen species regulate protrusion efficiency by iron oxide nanoparticles change endothelial cell morphology controlling actin dynamics,” PLoS One, vol. 7, no. 8, 2012. and mechanics via reactive oxygen species formation,” Jour- nal of Biomedical Materials Research Part A, vol. 96A, no. 1, [89] S. Esser, M. G. Lampugnani, M. Corada, E. Dejana, and pp. 186–195, 2011. W. Risau, “Vascular endothelial growth factor induces VE-cadherin tyrosine phosphorylation in endothelial cells,” [75] L. E. Crawford, E. E. Milliken, K. Irani et al., “Superoxide Journal of Cell Science, vol. 111, pp. 1853–1865, 1998. mediated actin response in post-hypoxic endothelial cells,” Journal of Biological Chemistry, vol. 271, no. 43, pp. 26863– [90] P. Andriopoulou, P. Navarro, A. Zanetti, M. G. Lampugnani, 26867, 1996. and E. Dejana, “Histamine induces tyrosine phosphorylation of endothelial cell-to-cell adherens junctions,” Arteriosclero- [76] L. Moldovan, N. I. Moldovan, R. H. Sohn, S. A. Parikh, and sis, Thrombosis, and Vascular Biology, vol. 19, no. 10, P. J. Goldschmidt-Clermont, “Redox changes of cultured pp. 2286–2297, 1999. endothelial cells and actin dynamics,” Circulation Research, vol. 86, no. 5, pp. 549–557, 2000. [91] F. E. Nwariaku, Z. Liu, X. Zhu et al., “NADPH oxidase medi- ates vascular endothelial cadherin phosphorylation and [77] A. Aghajanian, E. S. Wittchen, S. L. Campbell, and endothelial dysfunction,” Blood, vol. 104, no. 10, pp. 3214– K. Burridge, “Direct activation of RhoA by reactive oxygen 3220, 2004. species requires a redox-sensitive motif,” PLoS One, vol. 4, no. 11, article e8045, 2009. [92] R. E. Huie and S. Padmaja, “The reaction of no with superox- ide,” Free Radical Research Communications, vol. 18, no. 4, [78] K. M. Mohazzab, P. M. Kaminski, and M. S. Wolin, “NADH pp. 195–199, 1993. oxidoreductase is a major source of superoxide anion in bovine coronary artery endothelium,” American Journal of [93] S. Moncada, R. M. Palmer, and E. A. Higgs, “Nitric-oxide - Physiology-Heart and Circulatory Physiology, vol. 266, no. 6, physiology, pathophysiology, and pharmacology,” Pharma- pp. H2568–H2572, 1994. cological Reviews, vol. 43, no. 2, pp. 109–142, 1991. 14 Applied Bionics and Biomechanics [94] T. Walsh, T. Donnelly, and D. Lyons, “Impaired endothelial nitric oxide bioavailability: a common link between aging, hypertension, and atherogenesis?,” Journal of the American Geriatrics Society, vol. 57, no. 1, pp. 140–145, 2009. [95] U. Landmesser, S. Dikalov, S. R. Price et al., “Oxidation of tetrahydrobiopterin leads to uncoupling of endothelial cell nitric oxide synthase in hypertension,” Journal of Clinical Investigation, vol. 111, no. 8, pp. 1201–1209, 2003. [96] F. Cosentino, J. E. Barker, M. P. Brand et al., “Reactive oxygen species mediate endothelium-dependent relaxations in tetrahydrobiopterin-deficient mice,” Arteriosclerosis, Throm- bosis, and Vascular Biology, vol. 21, no. 4, pp. 496–502, [97] D. Stuehr, S. Pou, and G. M. Rosen, “Oxygen reduction by nitric-oxide synthases,” Journal of Biological Chemistry, vol. 276, no. 18, pp. 14533–14536, 2001. [98] I. V. Turko and F. Murad, “Protein nitration in cardiovascu- lar diseases,” Pharmacological Reviews, vol. 54, no. 4, pp. 619– 634, 2002. [99] C. A. Reinhart-King, “How matrix properties control the self- assembly and maintenance of tissues,” Annals of Biomedical Engineering, vol. 39, no. 7, pp. 1849–1856, 2011. [100] D. E. Discher, P. Janmey, and Y. L. Wang, “Tissue cells feel and respond to the stiffness of their substrate,” Science, vol. 310, no. 5751, pp. 1139–1143, 2005. [101] K. Roy, Y. Wu, J. L. Meitzler et al., “NADPH oxidases and cancer,” Clinical Science, vol. 128, no. 12, pp. 863–875, 2015. [102] V. Sosa, T. Moliné, R. Somoza, R. Paciucci, H. Kondoh, and M. E. LLeonart, “Oxidative stress and cancer: an overview,” Ageing Research Reviews, vol. 12, no. 1, pp. 376–390, 2013. [103] J. C. Kohn, D. W. Zhou, F. Bordeleau et al., “Cooperative effects of matrix stiffness and fluid shear stress on endothelial cell behavior,” Biophysical Journal, vol. 108, no. 3, pp. 471– 478, 2015. [104] P. A. Galie, A. van Oosten, C. S. Chen, and P. A. Janmey, “Application of multiple levels of fluid shear stress to endo- thelial cells plated on polyacrylamide gels,” Lab on a Chip, vol. 15, no. 4, pp. 1205–1212, 2015. [105] W. O. Carter, P. K. Narayanan, and J. P. Robinson, “Intracel- lular hydrogen peroxide and superoxide anion detection in endothelial cells,” Journal of Leukocyte Biology, vol. 55, no. 2, pp. 253–258, 1994. International Journal of Advances in Rotating Machinery Multimedia Journal of The Scientific Journal of Engineering World Journal Sensors Hindawi Hindawi Publishing Corporation Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 http://www www.hindawi.com .hindawi.com V Volume 2018 olume 2013 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Journal of Control Science and Engineering Advances in Civil Engineering Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Submit your manuscripts at www.hindawi.com Journal of Journal of Electrical and Computer Robotics Engineering Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 VLSI Design Advances in OptoElectronics International Journal of Modelling & Aerospace International Journal of Simulation Navigation and in Engineering Engineering Observation Hindawi Hindawi Hindawi Hindawi Volume 2018 Volume 2018 Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com www.hindawi.com www.hindawi.com Volume 2018 International Journal of Active and Passive International Journal of Antennas and Advances in Chemical Engineering Propagation Electronic Components Shock and Vibration Acoustics and Vibration Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Applied Bionics and Biomechanics Hindawi Publishing Corporation

Stiff Substrates Enhance Endothelial Oxidative Stress in Response to Protein Kinase C Activation

Loading next page...
 
/lp/hindawi-publishing-corporation/stiff-substrates-enhance-endothelial-oxidative-stress-in-response-to-jvM23o2d44
Publisher
Hindawi Publishing Corporation
Copyright
Copyright © 2019 Rebecca Lownes Urbano et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ISSN
1176-2322
eISSN
1754-2103
DOI
10.1155/2019/6578492
Publisher site
See Article on Publisher Site

Abstract

Hindawi Applied Bionics and Biomechanics Volume 2019, Article ID 6578492, 14 pages https://doi.org/10.1155/2019/6578492 Research Article Stiff Substrates Enhance Endothelial Oxidative Stress in Response to Protein Kinase C Activation 1 2 1,2 Rebecca Lownes Urbano, Swathi Swaminathan, and Alisa Morss Clyne Mechanical Engineering and Mechanics, Drexel University, Philadelphia, PA, USA Biomedical Engineering, Science and Health Systems, Drexel University, Philadelphia, PA, USA Correspondence should be addressed to Alisa Morss Clyne; asm67@drexel.edu Received 1 October 2018; Revised 28 January 2019; Accepted 19 February 2019; Published 14 April 2019 Academic Editor: Estefanía Peña Copyright © 2019 Rebecca Lownes Urbano et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Arterial stiffness, which increases with aging and hypertension, is an independent cardiovascular risk factor. While stiffer substrates are known to affect single endothelial cell morphology and migration, the effect of substrate stiffness on endothelial monolayer function is less understood. The objective of this study was to determine if substrate stiffness increased endothelial monolayer reactive oxygen species (ROS) in response to protein kinase C (PKC) activation and if this oxidative stress then impacted adherens junction integrity. Porcine aortic endothelial cells were cultured on varied stiffness polyacrylamide gels and treated with phorbol 12-myristate 13-acetate (PMA), which stimulates PKC and ROS without increasing actinomyosin contractility. PMA-treated endothelial cells on stiffer substrates increased ROS and adherens junction loss without increased contractility. ROS scavengers abrogated PMA effects on cell-cell junctions, with a more profound effect in cells on stiffer substrates. Finally, endothelial cells in aortae from elastin haploinsufficient mice (Eln+/-), which were stiffer than aortae from wild-type mice, showed decreased VE-cadherin colocalization with peripheral actin following PMA treatment. These data suggest that oxidative stress may be enhanced in endothelial cells in stiffer vessels, which could contribute to the association between arterial stiffness and cardiovascular disease. 1. Introduction (Eln+/-) mice had enhanced angiotensin-induced vasocon- striction and impaired endothelium-dependent vasodilation Due to the highly mechanical nature of the cardiovascular [13], although aortae from these same animals do not [16]. system, cardiovascular disease has long been accepted as both In human subjects, endothelial flow-mediated vasodilation was inversely correlated with aortic stiffness [17, 18]. Thus, a biomechanical and biochemical disease. Arterial stiffness, which increases with hypertension and aging among others, arterial stiffness alone may contribute to cardiovascular risk is an independent predictor of cardiovascular risk [1–3]. by altering critical endothelial functions. Arteries reversibly stiffen when smooth muscle cells contract However, cardiovascular risk factors rarely occur in isola- and irreversibly stiffen as elastin is degraded and collagen tion but rather cluster in certain individuals. Little is known increases [4–9]. Stiff arteries have long been known to con- about how arterial stiffness interacts with other cardiovascu- tribute to cardiovascular mortality by increasing cardiac lar risk factors such as diabetes and inflammation. Both afterload [10]; more recently, stiff arteries have also been hyperglycemia and inflammatory cytokines such as tumor shown to contribute to endothelial dysfunction, an initiat- necrosis factor-α (TNF-α) increase endothelial cell oxidative ing step in atherosclerosis [11–13]. In vitro, endothelial stress due to increased production and decreased scavenging monolayers on stiff polyacrylamide (PA) gels were more of reactive oxygen species (ROS). Elevated ROS have been permeable [12, 14, 15]. In animal models, endothelial per- implicated in both hypertension and atherosclerosis patho- meability was elevated in stiffened aortae from older mice genesis [19]. ROS are a family of highly reactive oxygen- [12]. Mesenteric arteries from elastin haploinsufficient containing molecules, including superoxide (O ), hydrogen 2 2 Applied Bionics and Biomechanics aorta was cut into four segments—two thoracic segments and peroxide (H O ), hydroxyl radical ( OH), and peroxynitrite 2 2 (ONOO ), which play an important role in many signaling two abdominal segments—producing four samples per aorta. pathways, such as cell proliferation, survival, and metabolism Samples were carefully mounted, the endothelium facing up, [20]. Superoxide is produced by the mitochondrial electron on a coverslip using Loctite 401 medical grade adhesive transport chain, as well as through protein kinase C- (PKC-) (Henkel) and submersed in PBS. The endothelium was induced NADPH oxidase upregulation and activation removed by gentle scraping with a cotton-tip applicator, [21–23]. NADPH oxidase assembly at the cell membrane based on a published protocol [31]. Subendothelial stiffness requires Rac, which is enhanced by substrate stiffness [24]. was determined by AFM using precalibrated cantilevers The effect of arterial stiffness on endothelial ROS produc- (spring constants between 0.10 and 0.17 N/m) with 10 μm tion in response to an external stimulus such as hyperglyce- spherical tips. Between three and nine indentations were mia or TNF-α has not yet been investigated. However, made at different locations along each sample. The force- these risk factors also activate many other endothelial cell sig- indentation curve for each indentation was fit to the Hertz naling pathways. Therefore, to isolate substrate stiffness model down to 200 nm indentation using a custom effects on endothelial ROS production, we used phorbol MATLAB code to produce a stiffness value [32]. Subendothe- 12-myristate 13-acetate (PMA), a PKC activator widely lial stiffness was calculated as the average of the individual used to stimulate ROS production in vitro. In previous stiffness values of each sample. studies, PMA treatment increased endothelial monolayer permeability but did not increase actinomyosin contractility, 2.3. Cell Culture and Polyacrylamide (PA) Gel Sample as measured by silicon substrate wrinkling, myosin light chain Preparation. Primary porcine aortic endothelial cells (PAEC) (MLC) phosphorylation, or MLC kinase activation [25, 26]. were isolated by the collagenase dispersion method and cul- Phorbol esters may instead induce barrier loss through tured in low glucose Dulbecco’s modified Eagle’s medium intermediate filament or actin cytoskeleton reorganization (DMEM, Corning) supplemented with 5% fetal bovine [27–29]. Thus, PMA enables investigation of substrate serum (FBS, HyClone), 1% glutamine, and 1% penicillin- stiffness effects on ROS production without stimulating streptomycin (Invitrogen). Cells were used up to passage 9. actinomyosin contractility. 6, 14, or 29 kPa was selected for the PA gel stiffnesses We hypothesized that stiff substrates would increase based on the subendothelial stiffnesses measured in WT endothelial ROS in response to PMA, resulting in actin fiber and Eln+/- mouse aorta (Figure 1). PA gels were prepared formation and cell-cell junction loss. We used varied stiffness following well-established protocols [33, 34]. Briefly, a bot- PA gels to study PMA-induced endothelial ROS, actin fiber tom coverslip was made hydrophilic by consecutive incuba- formation, and adherens junction loss. Abdominal aortae tions with 0.1 M sodium hydroxide (NaOH, Sigma-Aldrich), from wild-type (WT) and Eln+/- mice were treated with 3-aminopropyltrimethoxysiliane (3-APTES, Sigma-Aldrich), PMA ex vivo and imaged en face. We now show that sub- and 0.5% glutaraldehyde (Electron Microscopy Sciences). A strate stiffness enhances PMA-induced oxidative stress in top coverslip was made hydrophobic by applying SurfaSil endothelial monolayers in vitro and alters actin fiber reorga- (1,7-dichloro-octamethyltetrasiloxane, Thermo Scientific). nization and adherens junction morphology both in vitro A solution containing varying amounts of 40% acrylamide and ex vivo. and 2% bisacrylamide (Bio-Rad) was prepared based on the desired gel stiffness (Table 1). Ammonium persulfate (Bio-Rad) and tetramethylethylenediamine (TEMED, Bio- 2. Materials and Methods Rad) were added to the acrylamide/bisacrylamide solution 2.1. Animals. All experiments were performed according to to achieve final concentrations of 0.1% w/v and 0.3% v/v, respectively, initiating gel polymerization. Polymerizing gel protocols approved by the Drexel University College of Medicine Animal Studies Committee. Eln+/- mice were solution was added to the bottom coverslip, and the top generated as previously described [30]. 8-12-week-old coverslip was quickly inverted onto the polymerizing gel to create a flat surface. After gel formation, the top cover- WT and Eln+/- mice of both sexes, backcrossed several generations into the C57BL/6 background (Charles River), slip was removed. Elastic modulus was confirmed by AFM. To make the surface adhesive to cells, the gel was were used. All mice were genotyped to confirm elastin het- erozygosity, and decreased elastin lamellae thickness was UV-activated using sulfo-SANPAH (Thermo Fisher) in confirmed in select animals by immunohistochemistry. dimethyl sulfoxide (DMSO, Fisher Scientific) and 50 mM HEPES buffer and then incubated with 100 μg/mL type I Mice were provided access to food and water ad libitum ° ° at 22 C and a 12-hour light/dark cycle. collagen (BD Biosciences) at 37 C for 3 hours at room temperature or at 4 C overnight. The collagen-coated gel 2.2. Atomic Force Microscopy (AFM). AFM was used to quan- was rinsed in sterile phosphate-buffered saline (PBS) and tify aortic stiffness in wild-type (WT) and Eln+/- mouse aor- UV-sterilized prior to cell seeding. tae. The aorta was dissected and transferred to ice-cold PAEC were seeded on collagen-coated PA gels in phenol HEPES buffer (140 mM NaCl, 5 mM KCl, 1 mM CaCl , red-free DMEM and cultured to confluence for three days in 1.2 mM MgSO , 1.2 mM Na HPO , 10 mM HEPES, 10 mM a growth medium. Cells were then serum-starved overnight 4 2 4 sodium acetate, and 5 mM glucose, pH 7.4). Excess tissue in phenol red-free DMEM containing 1% FBS, 1% glutamine, was cleaned from the outside of the vessel, and the vessel and 1% penicillin-streptomycin. After serum starvation, cells was cut open longitudinally to expose the endothelium. Each were left untreated or treated with 1 μM PMA for varying Applied Bionics and Biomechanics 3 Thoracic Abdominal (a) (b) (c) Figure 1: Subendothelial stiffness increased in the thoracic and abdominal aortae of Eln+/- mice. (a) Longitudinally dissected mouse aorta opened to expose the endothelial surface. (b, top) Unscraped aorta showing the intact endothelium via β-catenin (green), cell structure using actin (red) and nuclei (blue), and the subendothelial matrix using collagen IV (white). Artery is shown en face. (b, bottom) Scraped aorta showed that the endothelium was removed since no β-catenin (green) was observed. The subendothelial matrix (collagen IV, white) remained intact and contiguous both en face and in cross section (smaller images). Scale bar = 50 μm. (c) Subendothelial stiffness of aortae from WT and Eln+/- mice. Thoracic and abdominal aortic sections were indented by AFM using a silicon nitride cantilever with a 10 μm spherical tip to measure subendothelial stiffness ( p <0 01 and p <0 05 by Student’s t-test). Three aortae were tested for each condition. Table 1: Acrylamide and bisacrylamide concentrations used to durations. In some cases, endothelial monolayers were pre- create varying elastic modulus PA gels. treated with ROS scavengers (4 mM N-acetyl cysteine or 50 mM sodium pyruvate, Sigma) for 1 hour prior to PMA. Acrylamide Bisacrylamide Elastic modulus 7.5% 0.05% 6 kPa 2.4. ROS Assay. ROS were measured using 5-(and-6)- 10% 0.1% 14 kPa ′ ′ chloromethyl-2 ,7 -dichlorodihydrofluorescein diacetate (CM-H DCFDA), which passively diffuses into cells where 10% 0.3% 29 kPa it is cleaved by intracellular esterases and then oxidized by Scraped Unscraped 4 Applied Bionics and Biomechanics ROS to yield a fluorescent adduct. 100 μM tert-butyl collected in prechilled Eppendorf tubes, and centrifuged at hydroperoxide (tBHP), which produces intracellular hydro- 4 C for 15 minutes at 13,000 rpm. The supernatant was col- gen peroxide, was the positive control. After PMA or tBHP lected, and the protein concentration was determined by treatment, samples were rinsed with warmed HBSS buffer BCA assay (Thermo Fisher). PKC activity was quantified in (0.137 M NaCl, 5.4 mM KCl, 0.25 mM Na HPO , 5.6 mM control or treated cell lysates using an ELISA-based PKC 2 4 glucose, 0.44 mM KH PO , 1.3 mM CaCl , 1.0 mM MgSO , kinase assay (Enzo) as per the manufacturer’s instructions. 2 4 2 4 and 4.2 mM NaHCO ). 25 μM CM-H DCFDA in phenol Absorbance (450 nm) was measured on a microplate reader 3 2 red-free DMEM was added to each sample and incubated (Thermo LabSystems, Multiskan Spectrum). Relative kinase for 25 minutes at 37 C, protected from light. To label nuclei, activity was calculated as follows: bisbenzimide (0.2 μg/mL, Thermo Fisher) was added to each sample for an additional 5 minutes. After thorough washing in HBSS buffer, samples were immersed in warmed phenol Relative kinase activity red-free DMEM and imaged in an Olympus Fluoview 1000 Average absorbance − Average absorbance sample blank microscope as confocal z-stacks (1 μm step size). = Quantity of crude protein used per assay 2.5. Confocal Microscopy and Image Analysis. Endothelial cells on PA gels were imaged by confocal microscopy and analyzed using MATLAB. In vitro cell samples were rinsed 2.7. Statistical Analysis. All statistical analyses were once with ice-cold PBS and fixed with ice-cold 4% parafor- conducted using MATLAB’s statistical toolbox. Graphs maldehyde. Samples were then permeabilized with 0.2% represent mean ± standard deviation. Multiple groups were Triton X-100 in PBS for 15 minutes and blocked with 1% compared using either two-way or n-way ANOVA with post bovine serum albumin (BSA) in PBS for 1 hour at room tem- hoc Tukey-Kramer test, and two groups were compared by perature. After fixation, mouse aortae were simultaneously Student’s t-test. Within each PKC assay, conditions were blocked and permeabilized in PBS containing 1% BSA and tested in duplicate. For ROS measurement, conditions were 0.3% Triton X-100. Endothelial cells on PA gels were labeled tested in triplicate. All experiments were conducted at least with primary antibodies for VE-cadherin (1 : 200, Santa two times, with at least three samples per condition. Cruz), β-catenin (1 : 200, Thermo Fisher), or pMLC (1 : 200, Cell Signaling) in 1% BSA in PBS overnight at 4 C. After sev- eral rinses with PBS, samples were then incubated with the 3. Results appropriate Alexa Fluor 488 or 633 (1 : 200, Thermo Fisher) secondary antibody, rhodamine phalloidin (16.5 nM, 3.1. Subendothelial Stiffness Was Higher in Eln+/- as Invitrogen), and bisbenzimide (0.2 μg/mL) for 1 hour at Compared to WT Mouse Aorta. Macroscale arterial stiffness, room temperature, protected from light. Samples were rinsed measured by pulse wave velocity or pressure myography, twice with PBS and mounted in 1 : 1 glycerol : PBS. Confocal increases in mice genetically engineered to produce less z-stacks were acquired for all samples with either a 0.25 or elastin (Eln+/-) [36, 37]. However, aortic stiffness had 0.5 μm step size (for in vitro and ex vivo samples, respec- not been characterized by atomic force microscopy in this tively) using an Olympus Fluoview 1000 confocal micro- mouse model. The longitudinally dissected mouse aorta scope at 60x magnification. (Figure 1(a)) shows the mounting technique and the thoracic A custom MATLAB code was created to quantify ROS and abdominal sections. Aortae with the intact endothelium and pMLC. The background was subtracted using a 50 × 50 were first labeled for β-catenin (green) to show endothelial pixel area. Images were then binarized using the same thresh- cell-cell junctions, actin (red) and nuclei (blue) to highlight old as determined using Otsu’s method, which calculates a cell structure, and collagen IV (white) to view the basement threshold based on pixel intensity distribution [35]. Noise membrane (Figure 1(b), top). When we removed the endo- was removed from binarized images by excluding small thelium from the longitudinally dissected mouse aortae, we objects (less than 9 pixels for ROS, less than 30 pixels for no longer observed β-catenin, confirming that cells were pMLC). The number of remaining pixels with intensities removed. The collagen IV layer remained intact and contig- above the threshold (“positive” pixels) was counted for uous, indicating that the subendothelial matrix remained each image. Three images per sample were quantified intact (Figure 1(b), bottom; collagen IV integrity was most using this method and averaged to quantify ROS or pMLC clear in the cross-sectional image). However, it is possible in each sample. that endothelial removal did damage the subendothelial layer, as evidenced by the spaces in the fluorescently labeled 2.6. PKC Activity Assay. After treatment, cells on PA gels samples. We therefore repeated the atomic force microscopy were quickly rinsed with ice-cold PBS and inverted onto in both scraped and unscraped WT mouse aortae and found 50 μL lysis buffer (20 mM MOPS, 50 mM β-glycerophos- no difference in aortic stiffness measurements. When aortic phate, 50 mM sodium fluoride, 1 mM sodium orthovanadate, samples were indented by AFM, the thoracic and abdominal 5 mM EGTA, 2 mM EDTA, 1% NP40, 1 mM dithiothreitol, aortic subendothelium from Eln+/- mice was about 1.75-fold 1 mM benzamidine, 1 mM phenylmethanesulfonyl fluoride, stiffer than that of WT mice (Figure 1(c)). The thoracic 10 μg/mL leupeptin, and 10 μg/mL aprotinin) for 10 minutes aorta was consistently stiffer than the abdominal aorta in at 4 C. Lysed cells were then scraped from the gel substrates, both genotypes. Applied Bionics and Biomechanics 5 Untreated PMA tBHP Nuclei ROS Nuclei ROS Nuclei ROS 2.0 1.5 1.0 0.5 0.0 6 kPa 14 kPa 29 kPa Substrate stiffness Figure 2: ROS increased with substrate stiffness following PMA treatment. PAEC monolayers on 6, 14, and 29 kPa gels were treated with 1 μM PMA for 10 minutes. Tert-butyl hydroperoxide (tBHP) was the positive control. Cell nuclei were labeled with Hoechst, and ROS with CM-H DCFDA. Samples were imaged at 20x by confocal microscopy. Scale bar is 25 μm. Oxidative stress was quantified using the number of positive pixels (above the threshold) using the custom MATLAB code. PMA-treated samples were normalized to untreated samples on the same substrate stiffness. The effect of substrate stiffness was significant by one-way ANOVA (p <0 01). p <0 05 and p <0 01 by post hoc Tukey-Kramer test. 3.2. PMA-Induced Increased Oxidative Stress in Endothelial significantly whether the cells were on soft or stiff PA gels Cells on the Stiffest PA Gels. We then created 6, 14, and (Figure 3(b)). Therefore, the PMA-induced differences in 29 kPa PA gels, which correspond to aortic stiffnesses in oxidative stress on stiffer substrates were not related to WT and Eln+/- mice, to determine if different ROS levels PKC activation. were produced by PMA-treated endothelial monolayers cultured on substrates of varied stiffness. Each sample was 3.4. Endothelial Cells Formed More Actin Stress Fibers in treated with 1 μM PMA for 10 minutes, based on preliminary Response to PMA on Stiffer Gels. ROS lead to endothelial experiments showing maximum cell viability and ROS at this actin fiber formation [40]. PMA-stimulated cells on increas- dose and time, and consistently imaged by confocal micros- ing stiffness substrates were labeled for actin fibers to copy. ROS were statistically similar following PMA treatment determine if substrate stiffness-dependent oxidative stress of cells on 6 and 14 kPa gels. In addition, endothelial cells on increased actin fiber formation. In untreated samples, actin stiff substrates did not show any baseline increase in ROS. fibers were primarily located around the cell periphery on However, in endothelial cells on 29 kPa substrates that were all substrates, although the effect was more pronounced in treated with PMA, ROS increased by more than 50% cells on the softest 6 kPa gels (Figure 4(a), representative cell (p <0 magnification in Figure 4(b)). Following 15 minutes of PMA 01 compared to fold change in cells on 6 kPa gels, Figure 2). Substrate stiffness effects on the PMA-induced fold treatment, actin fibers appeared in cells on the 14 and 29 kPa change in ROS were also significant by one-way ANOVA gels, but not in cells on the 6 kPa gels. This effect was even (p <0 01). These results demonstrate that stiffer substrates more pronounced following 30 minutes of PMA treatment, increase endothelial ROS in response to PMA. with larger stress fibers and nearly complete peripheral actin loss in endothelial cells on 14 and 29 kPa gels. Cells on 6 kPa 3.3. Endothelial Cell PKC Increased in Response to PMA gels largely retained peripheral actin with PMA treatment. Independent of PA Gel Stiffness. PMA induces ROS produc- PAECs were then labeled for pMLC to determine tion through PKC signaling [38, 39]. We therefore measured whether ROS-induced actin stress fiber formation was PKC activity in PMA-treated endothelial cells on 6, 14, and associated with increased actinomyosin contractility through pMLC localization to actin stress fibers. 1 μM PMA treat- 29 kPa PA gels to determine if PKC activation increased on stiffer substrates. PKC activity in PAEC increased 3-4-fold ment for 15 or 30 minutes did not induce pMLC transloca- within 5 minutes of PMA treatment (Figure 3(a)). However, tion to actin fibers or increase overall pMLC (Figure 5). In PKC activity in cells stimulated with PMA did not change contrast, the positive control (10 U/mL thrombin for 30 29 kPa 14 kPa 6 kPa Fold change in ROS (PMA/untreated) 6 Applied Bionics and Biomechanics 1.5 1.0 # ⁎ 휇 0.5 0 0.0 Untreated PMA 6 kPa 14 kPa 29 kPa (a) (b) Figure 3: PKC activity increased with PMA independent of substrate stiffness. (a) PKC activity was measured following 1 μM PMA treatment in endothelial cells on glass substrates. Purified active PKC was the positive control. p <0 01 compared to that untreated by Student’s t-test. (b) PAEC monolayers on 6, 14, and 29 kPa substrates were treated with 1 μM PMA for 5 minutes. PMA was significant by two-way ANOVA (p <0 001), but substrate stiffness was not significant. p <0 05 by post hoc Tukey-Kramer test compared to that untreated. Untreated 15 min PMA 30 min PMA Untreated 15 min PMA 30 min PMA (a) (b) Figure 4: Actin stress fiber formation was greater in endothelial cells on stiffer substrates following PMA. PAEC monolayers on 6, 14, or 29 kPa gels were treated with 1 μM PMA for 15 or 30 minutes prior to fixation and immunofluorescent labeling of actin (rhodamine phalloidin). (a) Maximum intensity projection from confocal z-stacks at 60x magnification. Scale bar is 25 μm. (b) Magnified representative cells from (a). minutes) increased overall pMLC approximately 14-fold, 3.5. Adherens Junctions Became Less Reticular in Response to with pMLC localized along the actin fibers. These results PMA on Stiffer Gels. ROS induce cell-cell junction loss, which indicate that PMA-induced actin fiber formation in cells has been attributed in part to adherens junction protein phos- on stiffer substrate did not correspond to actinomyosin phorylation and internalization [41–43]. We therefore mea- contractility. sured if stiff substrates exacerbate ROS-mediated endothelial 29 kPa 14 kPa 6 kPa Absorbance (a.u.) Blank Active PKC Untreated PMA, 5 min PMA, 10 min 29 kPa 14 kPa 6 kPa Normalized PKC activity Absorbance (a.u./g protein) Applied Bionics and Biomechanics 7 Untreated 15 min PMA 30 min PMA 6 kPa 14 kPa 29 kPa Substrate stiffness Untreated 15 min PMA 30 min PMA Thrombin Thrombin (positive control) Figure 5: PMA treatment did not increase pMLC localization to actin stress fibers in cells on varied stiffness substrates. PAEC monolayers on 6, 14, or 29 kPa gels were treated with 1 μM PMA for 15 or 30 minutes, prior to fixation and pMLC immunofluorescent labeling. For the positive control, cells on a 29 kPa gel were treated with 10 U/mL thrombin for 30 minutes. Images are maximum intensity projections from 60x confocal z-stacks. Scale bar is 25 μm. pMLC-positive pixels were quantified using the custom MATLAB code. Stiffness and PMA treatment were not significant by n-way ANOVA. Untreated 15 min PMA 30 min PMA Untreated 15 min PMA 30 min PMA (a) (b) Figure 6: Reticular adherens junction loss was greater in cells on stiffer substrates following PMA. PAEC monolayers on 6, 14, or 29 kPa gels were treated with 1 μM PMA for 15 or 30 minutes, prior to fixation and immunofluorescent labeling of the cell-cell junction protein β-catenin. (a) Maximum intensity projection from confocal z-stacks at 60x magnification. Scale bar is 25 μm. (b) Representative junctions. adherens junction loss in response to PMA. In untreated representative magnified junctions in Figure 6(b)). In con- cells on 6, 14, and 29 kPa gels, wide reticular adherens junc- trast to what has been observed in other published work, tions were evident between adjacent cells (Figure 6(a), we did not observe any changes in adherens junction based 29 kPa 14 kPa 6 kPa 29 kPa 14 kPa 6 kPa 29 kPa 14 kPa 6 kPa pMLC-positive (thousands) 8 Applied Bionics and Biomechanics 훽-Catenin Untreated PMA (a) (b) 훽-Catenin Actin Untreated PMA Untreated PMA (c) (d) Figure 7: ROS scavengers prevented PMA-induced adherens junction loss and actin fiber redistribution. PAEC monolayers were pretreated with ROS scavengers (4 mM N-acetyl cysteine, 50 mM sodium pyruvate) for 1 hour before the 30-minute treatment with 1 μM PMA. Samples were fixed and immunofluorescently labeled for (a) β-catenin, with representative cells magnified in (c), and immunofluorescently labeled for (b) actin, with representative cells magnified in (d). Images are maximum intensity projections from 60x confocal z-stacks. Scale bar is 25 μm. on substrate stiffness alone perhaps due to the use of a differ- junction morphology change was abrogated. More strikingly, ent endothelial cell type [12]. After 15 or 30 minutes of PMA ROS scavengers prevented PMA-induced actin reorgani- treatment, reticular junctions were mostly maintained in cells zation (Figure 7(b), representative cell magnification in on the 6 kPa gels. In contrast, cells on the stiffest 29 kPa sub- Figure 7(d)). Cells treated with the ROS scavengers prior strates lost most junction reticular structures and instead had to PMA showed peripheral actin bands which were similar to linear or disrupted cell-cell junctions. These results demon- those in untreated cells. Thus, oxidative stress was likely responsible for PMA-induced reticular junction loss and strate that endothelial reticular junction structure loss is exacerbated by stiffer substrates in response to the ROS- actin fiber formation. stimulant PMA. To confirm that ROS were responsible for the PMA- 4. Discussion induced changes in adherens junctions and actin fiber formation, the ROS scavengers N-acetyl cysteine and sodium Oxidative stress and more specifically the enzyme responsi- pyruvate were administered for 1 hour prior to PMA treat- ble for superoxide production, NADPH oxidase, have been ment. These experiments were performed on glass coverslips, implicated in cardiovascular disease and atherosclerosis in since junction loss following PMA exposure was highest on particular [44, 45]. We now show that stiffer substrates exac- these stiffest substrates. ROS scavengers themselves did not erbate endothelial cell oxidative stress. In response to PMA, affect cell-cell junction structure, and cells treated with endothelial cells on the stiffest substrates showed more ROS PMA alone showed linear and invaginated adherens junctions and actin stress fibers and showed greater adherens junction (Figure 7(a), representative cell magnification in Figure 7(c)). loss, which was not attributed to cell contractility. Stiffer In PAEC pretreated with ROS scavengers prior to PMA, the aortas from Eln+/- mice also showed less VE-cadherin at ROS Scavenger None ROS Scavenger None ROS Scavenger None Applied Bionics and Biomechanics 9 stiffness. However, while these morphological responses cell-cell membranes and increased peripheral actin fiber formation in response to PMA. Since PMA-induced PKC to substrate stiffness are no longer observed as endothelial activation was not affected by substrate stiffness, it is likely cells reach confluency, this study shows that both endothe- that substrate stiffness affected cells through alternative lial biochemical responses and cell-cell interactions do pathways. These data suggest that oxidative stress and its det- change with substrate stiffness. rimental downstream effects on endothelial cells and vascular ROS, specifically superoxide and its byproduct hydrogen disease may be enhanced in stiffer arteries. peroxide, have been shown to regulate actin fibers in vascular The vascular mechanics of the elastin haploinsufficient cells [40, 72–74]. Actin fiber formation in subconfluent mouse have been extensively studied, both in terms of passive reoxygenated hypoxic aortic endothelial cells was inhibited mechanical stretch in response to increasing pressure and in by superoxide dismutase overexpression, suggesting a key terms of vasodilation and constriction in response to role for superoxide [75]. Superoxide can reversibly oxidize biochemical stimuli [13, 37, 46, 47]. These studies focused proteins, including actin itself. In endothelial cells, actin primarily on the decrease in total elastin in the vascular wall, oxidation may be essential for actin polymerization during as well as the increase in elastin lamellae. We and others did cell migration. For example, in migrating mouse aortic endo- not find any changes in other extracellular matrix proteins, in thelial cells, actin monomer incorporation into actin fibers particular collagen, which is the other primary protein was diminished following treatment with the NADPH oxi- thought to define vascular wall stiffness. However, some dase inhibitor diphenyleneiodonium (DPI) and a superoxide recent studies in other tissues in elastin haploinsufficient mice dismutase mimetic [76]. Alternatively, superoxide can oxi- suggest that there are also collagen changes in these animals. dize RhoA, enhancing GDP dissociation and subsequent The lungs of Eln +/- mice contained nearly twice as much col- activation. Fibroblasts with an oxidation-resistant form of lagen 1 and lysyl oxidase, an important collagen crosslinker, as RhoA did not form stress fibers in response to hydrogen WT mice [48]. Achilles tendons of Eln+/- mice had the same peroxide [77]. While the source of increased ROS in endo- total collagen content as WT mice but different collagen fibril thelial cells on stiffer substrates remains unknown, we diameter distribution [49]. Thus, it is possible that the hypothesize that stiff substrates increase NADPH oxidase increased stiffness we measured in the Eln+/- aorta relates production or assembly, since NADPH oxidase appears to changes in collagen content and/or structure. to produce the most nonmitochondrial superoxide in The thoracic aorta was consistently stiffer than the endothelial cells [78]. We hope to investigate this mechanism abdominal aorta in both WT and Eln+/- mice. These data further in future studies. agree with human studies in which aortic stiffness decreased Although endothelial oxidative stress has not been with distance from the heart, especially in older patients studied on varied stiffness substrates, endothelial superoxide [50, 51]. Other studies in C57BL/6 mice demonstrated that production is mechanosensitive, specifically in response to the aortic elastic modulus was highest in the distal thoracic shear stress [79]. Bovine aortic endothelial cells produced aorta and lowest in the abdominal aorta [52]. In a subsequent three times more superoxide under oscillatory shear stress study, it was shown that the decrease in aortic stiffness along compared to laminar flow [80]. Shear stress activates Rac, the length of the aorta was accounted for by a decrease in which is downstream of integrin activation and contributes total and lamellar elastin [53]. Since our data show a propor- to ROS production [81, 82]. Epithelial cells have been shown tionally similar decrease in aortic stiffness from the thoracic to produce more ROS when on stiffer substrates. MMP-3- to the abdominal sections in both WT and Eln+/- mice, stimulated ROS production was approximately 3.5-fold higher in epithelial cells on 4.02 kPa substrates compared to it is likely that elastin content is important to the regional stiffness variation. 0.13 kPasubstrates;β1integrinsubunitknockdowneliminated Our data support other recently published studies show- ROSproductioninresponsetoMMP-3[83].These findings sug- ing that substrate stiffness affects not only single endothelial gest that integrin activation-induced Rac1 activity contrib- cells but also confluent endothelial monolayers [12]. In vitro utes to ROS production in cells on stiffer substrates [84]. studies of cell response to substrate stiffness began when The increase in adherens junction disruption could be Pelham and Wang first used protein-coated PA gels to show either contractility-dependent or contractility-independent. that both rat kidney epithelial and 3T3 fibroblasts spread Permeability agents, including thrombin, lipopolysaccha- to a greater extent on stiff than soft substrates [54]. Since ride (LPS), TNF-α, and vascular endothelial growth factor that seminal paper, many cell types were shown to change (VEGF), activate the Rho/ROCK pathway and cell con- their morphology [55–58], motility [59, 60], differentiation tractility [85]. The ROCK inhibitor Y-27632 prevented [61, 62], and proliferation [63, 64] in response to substrate adherens junction disruption in endothelial monolayer stiffness. For endothelial cells specifically, single cells increase studies, although some effects may be endothelial the spread area [65, 66], stiffness [67], cell-matrix and cell- subtype-dependent (e.g., macrovascular or microvascular) cell forces [66, 68], and proliferation [69] with substrate stiff- [86, 87]. In epithelial cell protrusions, H O increased 2 2 ness. However, as cells proliferated and reached confluency, actin polymerization, cofilin activity, and barbed ends; substrate-dependent differences were diminished or no however, myosin IIA did not colocalize with actin fibers in longer observed [70, 71]. Similarly, we did not observe H O -treated cell protrusions [88]. These data fit with our 2 2 any changes in endothelial cell and actin stress fiber mor- results that actin contractility did not increase with oxidative phology, focal adhesion size, or focal adhesion number in stress. Therefore, it is more likely in our case that oxidative endothelial cell monolayers on substrates of different stress induced contractility-independent adherens junction 10 Applied Bionics and Biomechanics disruption. ROS also disrupt cell-cell junctions through 5. Conclusions VE-cadherin phosphorylation. Endothelial cell treatment This work illustrates a novel potential mechanism for with permeability agonists, such as VEGF and histamine, substrate-enhanced oxidative stress in response to PKC resulted in VE-cadherin tyrosine phosphorylation [89, 90]. activation in the endothelium. Since many endothelial cell In HUVECs, the ROS scavenger N-acetyl cysteine prevented studies are performed on tissue culture polystyrene of VE-cadherin phosphorylation by TNF-α [91]. Thus, we essentially infinite stiffness, these studies may overestimate hypothesize that adherens junction protein phosphorylation endothelial cell response to stressors. Further study of the resulted in cell-cell junction loss, although we did not directly interaction between arterial stiffness and oxidative stress measure it. could improve therapies to prevent or ameliorate endothelial ROS can also lead to adverse effects on the endothelium barrier dysfunction. beyond adherens junction loss. For example, superoxide (O ) interacts with nitric oxide (NO) to form peroxynitrite. Data Availability This interaction effectively decreases the NO availability, which is needed for vasodilation [92, 93]. Superoxide also AFM indentation curves, PKC quantification, and immuno- uncouples eNOS by oxidizing the eNOS cofactor tetrahydro- fluorescence images used to support the finding of this study biopterin (BH )[94–96]. Uncoupled eNOS produces more may be released upon application to the corresponding superoxide instead of NO [97], which further increases author. peroxynitrite. Protein nitration by peroxynitrite is widely observed in many cardiovascular diseases [98]. Thus, vascu- Conflicts of Interest lar stiffness-induced endothelial oxidative stress could have damaging effects beyond endothelial barrier function. The authors declare that there is no conflict of interest Substrate stiffness also affects other cell types beyond regarding the publication of this paper. endothelial cells, including fibroblasts, breast cancer cells, and stem cells [99, 100]. In vivo, tumors are stiffer than their Authors’ Contributions surrounding environment, which may alter both basal func- tion and inflammatory response in all of these cell types. In RLU helped design the study, carried out all experiments, addition, some tumors overexpress specific NADPH oxidases helped analyze the data, and drafted the manuscript. SS [101]. This overexpression could couple with increased tumor assisted with data analysis. AMC helped design the study, stiffness to further magnify oxidative stress in tumors. Tumor analyzed the data, and drafted the manuscript. All authors oxidative stress contributes to tissue injury and DNA damage gave final approval for publication. in premalignant conditions, as well as in cancer initiation and progression. Since the tumor cells themselves may be resistant Acknowledgments to oxidative stress, the injury to the surrounding tissue may be more severe [102]. Thus, stiffness-associated ROS inhibition This work was supported by the American Heart Association could potentially decrease cancer-induced damage and (grant number SDG4460068). We thank Patrick Osei-Owusu inhibit cancer metastasis through compromised blood vessels. for providing the WT and Eln+/- mouse aortae. While our work shows that PMA-induced oxidative stress increases with substrate stiffness, our research is not References without limitations. Phorbol esters, including PMA, are found in certain plants and can cause toxicity in animals [1] S. Laurent, J. Cockcroft, L. van Bortel et al., “Expert consensus when consumed [38]. Yet, PMA itself is not involved in document on arterial stiffness: methodological issues and clinical applications,” European Heart Journal, vol. 27, cardiovascular disease pathogenesis. We chose to use PMA no. 21, pp. 2588–2605, 2006. to isolate ROS production from cell contractility in cell-cell [2] C. Vlachopoulos, K. Aznaouridis, and C. Stefanadis, “Predic- junction loss [25, 26]; however, these studies should be tion of cardiovascular events and all-cause mortality with repeated using a physiologically relevant ROS inducer arterial stiffness: a systematic review and meta-analysis,” (e.g., tumor necrosis factor-α). PA gels do not fully capture Journal of the American College of Cardiology, vol. 55, the endothelial mechanical environment, including substrate no. 13, pp. 1318–1327, 2010. viscoelasticity and relaxation as well as shear stress or strain. [3] Y. Ben-Shlomo, M. Spears, C. Boustred et al., “Aortic pulse Several recent papers have shown interactions between shear wave velocity improves cardiovascular event prediction: an stress and substrate stiffness, demonstrating that softer sub- individual participant meta-analysis of prospective observa- strates augmented the atheroprotective effects of laminar tional data from 17,635 subjects,” Journal of the American shear stress [103, 104]. We also were not able to specifically College of Cardiology, vol. 63, no. 7, pp. 636–646, 2014. detect superoxide in intact vessels under physiological [4] A. J. Bank, H. Wang, J. E. Holte, K. Mullen, R. Shammas, and conditions due to technical challenges. In future studies, S. H. Kubo, “Contribution of collagen, elastin, and smooth a superoxide-specific indicator such as dihydroethidium muscle to in vivo human brachial artery wall stress and elastic could be used along with NADPH oxidase component knock- modulus,” Circulation, vol. 94, no. 12, pp. 3263–3270, 1996. down to support the hypothesis that stiffness-induced integ- [5] B. S. Fleenor, “Large elastic artery stiffness with aging: novel rin activation enhances superoxide production by NADPH translational mechanisms and interventions,” Aging and oxidase [105]. Disease, vol. 4, no. 2, pp. 76–83, 2013. Applied Bionics and Biomechanics 11 [21] F. Cosentino, M. Eto, P. de Paolis et al., “High glucose causes [6] M. Veerasamy, G. A. Ford, D. Neely et al., “Association of aging, arterial stiffness, and cardiovascular disease: a review,” upregulation of cyclooxygenase-2 and alters prostanoid Cardiology in Review, vol. 22, no. 5, pp. 223–232, 2014. profile in human endothelial cells: role of protein kinase C and reactive oxygen species,” Circulation, vol. 107, no. 7, [7] C. Xu, C. K. Zarins, P. S. Pannaraj, H. S. Bassiouny, and pp. 1017–1023, 2003. S. Glagov, “Hypercholesterolemia superimposed by experi- mental hypertension induces differential distribution of [22] T. Inoguchi, P. Li, F. Umeda et al., “High glucose level and collagen and elastin,” Arteriosclerosis, Thrombosis, and free fatty acid stimulate reactive oxygen species production Vascular Biology, vol. 20, no. 12, pp. 2566–2572, 2000. through protein kinase C–dependent activation of NAD(P)H oxidase in cultured vascular cells,” Diabetes, vol. 49, no. 11, [8] M. Cecelja and P. Chowienczyk, “Dissociation of aortic pulse pp. 1939–1945, 2000. wave velocity with risk factors for cardiovascular disease other than hypertension: a systematic review,” Hypertension, [23] L. Quagliaro, L. Piconi, R. Assaloni, L. Martinelli, E. Motz, vol. 54, no. 6, pp. 1328–1336, 2009. and A. Ceriello, “Intermittent high glucose enhances apoptosis related to oxidative stress in human umbilical vein [9] A. Scuteri, P. G. Cunha, E. A. Rosei et al., “Arterial stiffness endothelial cells: the role of protein kinase C and NAD(P)H- and influences of the metabolic syndrome: a cross-countries oxidase activation,” Diabetes, vol. 52, no. 11, pp. 2795–2804, study,” Atherosclerosis, vol. 233, no. 2, pp. 654–660, 2014. [10] G. F. Mitchell, “Arterial stiffness: insights from Framingham [24] M. T. Elnakish, H. H. Hassanain, P. M. Janssen, M. G. and Iceland,” Current Opinion in Nephrology and Hyperten- Angelos, and M. Khan, “Emerging role of oxidative stress sion, vol. 24, no. 1, pp. 1–7, 2015. in metabolic syndrome and cardiovascular diseases: impor- [11] B. M. Kaess, J. Rong, M. G. Larson et al., “Aortic stiffness, tant role of Rac/NADPH oxidase,” The Journal of Pathol- blood pressure progression, and incident hypertension,” ogy, vol. 231, no. 3, pp. 290–300, 2013. Journal of the American Medical Association, vol. 308, no. 9, [25] J. G. N. Garcia, H. W. Davis, and C. E. Patterson, “Regulation pp. 875–881, 2012. of endothelial cell gap formation and barrier dysfunction: [12] J. Huynh, N. Nishimura, K. Rana et al., “Age-related intimal role of myosin light chain phosphorylation,” Journal of Cellu- stiffening enhances endothelial permeability and leukocyte lar Physiology, vol. 163, no. 3, pp. 510–522, 1995. transmigration,” Science Translational Medicine, vol. 3, [26] N. V. Bogatcheva, A. D. Verin, P. Wang et al., “Phorbol esters no. 112, article 112ra122, 2011. increase MLC phosphorylation and actin remodeling in [13] P. Osei-Owusu, R. H. Knutsen, B. A. Kozel, H. H. Dietrich, bovine lung endothelium without increased contraction,” K. J. Blumer, and R. P. Mecham, “Altered reactivity of resis- American Journal of Physiology-Lung Cellular and Molecular tance vasculature contributes to hypertension in elastin insuf- Physiology, vol. 285, no. 2, pp. L415–L426, 2003. ficiency,” American Journal of Physiology-Heart and [27] A. B. Moy, K. Blackwell, N. Wang et al., “Phorbol Circulatory Physiology, vol. 306, no. 5, pp. H654–H666, 2014. ester-mediated pulmonary artery endothelial barrier dys- [14] A. A. Birukova, X. Tian, I. Cokic, Y. Beckham, M. L. Gardel, function through regulation of actin cytoskeletal mechanics,” and K. G. Birukov, “Endothelial barrier disruption and recov- American Journal of Physiology-Lung Cellular and Molecular ery is controlled by substrate stiffness,” Microvascular Physiology, vol. 287, no. 1, pp. L153–L167, 2004. Research, vol. 87, pp. 50–57, 2013. [28] J. E. Stasek, C. E. Patterson, and J. G. N. Garcia, “Protein- [15] R. Krishnan, D. D. Klumpers, C. Y. Park et al., “Substrate kinase-c phosphorylates caldesmon77 and vimentin and stiffening promotes endothelial monolayer disruption enhances albumin permeability across cultured bovine through enhanced physical forces,” American Journal of pulmonary-artery endothelial-cell monolayers,” Journal of Physiology-Cell Physiology, vol. 300, no. 1, pp. C146–C154, Cellular Physiology, vol. 153, no. 1, pp. 62–75, 1992. [29] N. V. Bogatcheva, A. Birukova, T. Borbiev et al., “Caldesmon [16] G. Faury, M. Pezet, R. H. Knutsen et al., “Developmental is a cytoskeletal target for PKC in endothelium,” Journal of adaptation of the mouse cardiovascular system to elastin Cellular Biochemistry, vol. 99, no. 6, pp. 1593–1605, 2006. haploinsufficiency,” Journal of Clinical Investigation, vol. 112, no. 9, pp. 1419–1428, 2003. [30] A. J. Oliveira-dos-Santos, G. Matsumoto, B. E. Snow et al., “Regulation of T cell activation, anxiety, and male aggression [17] C. M. McEniery, S. Wallace, I. S. Mackenzie et al., “Endothe- by RGS2,” Proceedings of the National Academy of Sciences of lial function is associated with pulse pressure, pulse wave the United States of America, vol. 97, no. 22, pp. 12272– velocity, and augmentation index in healthy humans,” Hyper- 12277, 2000. tension, vol. 48, no. 4, pp. 602–608, 2006. [31] J. Peloquin, J. Huynh, R. M. Williams, and C. A. Reinhart- [18] A. Nigam, G. F. Mitchell, J. Lambert, and J. C. Tardif, “Rela- King, “Indentation measurements of the subendothelial tion between conduit vessel stiffness (assessed by tonometry) matrix in bovine carotid arteries,” Journal of Biomechanics, and endothelial function (assessed by flow-mediated dilata- vol. 44, no. 5, pp. 815–821, 2011. tion) in patients with and without coronary heart disease,” The American Journal of Cardiology, vol. 92, no. 4, pp. 395– [32] J. Solon, I. Levental, K. Sengupta, P. C. Georges, and P. A. Janmey, “Fibroblast adaptation and stiffness matching to 399, 2003. soft elastic substrates,” Biophysical Journal, vol. 93, no. 12, [19] A. M. Briones and R. M. Touyz, “Oxidative stress and hyper- pp. 4453–4461, 2007. tension: current concepts,” Current Hypertension Reports, vol. 12, no. 2, pp. 135–142, 2010. [33] J. R. Tse and A. J. Engler, “Preparation of hydrogel substrates with tunable mechanical properties,” Current Protocols in [20] P. D. Ray, B.-W. Huang, and Y. Tsuji, “Reactive oxygen Cell Biology, vol. 47, no. 1, pp. 10.16.1–10.16.16, 2010. species (ROS) homeostasis and redox regulation in cellular signaling,” Cellular Signalling, vol. 24, no. 5, pp. 981–990, [34] R. L. Urbano, C. Furia, S. Basehore, and A. M. Clyne, “Stiff 2012. substrates increase inflammation-induced endothelial 12 Applied Bionics and Biomechanics monolayer tension and permeability,” Biophysical Journal, [49] J. D. Eekhoff, F. Fang, L. G. Kahan et al., “Functionally dis- vol. 113, no. 3, pp. 645–655, 2017. tinct tendons from elastin haploinsufficient mice exhibit mild stiffening and tendon-specific structural alteration,” Journal [35] N. Otsu, “A threshold selection method from gray-level of Biomechanical Engineering, vol. 139, no. 11, p. 111003, histograms,” IEEE Transactions on Systems, Man, and Cyber- netics, vol. 9, no. 1, pp. 62–66, 1979. [50] V. Taviani, S. S. Hickson, C. J. Hardy et al., “Age-related [36] J. E. Wagenseil, C. H. Ciliberto, R. H. Knutsen, M. A. Levy, changes of regional pulse wave velocity in the descending A. Kovacs, and R. P. Mecham, “Reduced vessel elasticity aorta using Fourier velocity encoded M-mode,” Magnetic alters cardiovascular structure and function in newborn Resonance in Medicine, vol. 65, no. 1, pp. 261–268, 2011. mice,” Circulation Research, vol. 104, no. 10, pp. 1217–1224, 2009. [51] W. J. Rogers, Y. L. Hu, D. Coast et al., “Age-associated changes in regional aortic pulse wave velocity,” Journal of [37] L. Carta, J. E. Wagenseil, R. H. Knutsen et al., “Discrete con- the American College of Cardiology, vol. 38, no. 4, pp. 1123– tributions of elastic fiber components to arterial development 1129, 2001. and mechanical compliance,” Arteriosclerosis, Thrombosis, [52] X. Guo and G. S. Kassab, “Variation of mechanical properties and Vascular Biology, vol. 29, no. 12, pp. 2083–2089, 2009. along the length of the aorta in C57bl/6 mice,” American [38] G. Goel, H. P. S. Makkar, G. Francis, and K. Becker, “Phorbol Journal of Physiology-Heart and Circulatory Physiology, esters: structure, biological activity, and toxicity in animals,” vol. 285, no. 6, pp. H2614–H2622, 2003. International Journal of Toxicology, vol. 26, no. 4, pp. 279– [53] T. E. Clark, M. A. Lillie, A. W. Vogl, J. M. Gosline, and R. E. 288, 2007. Shadwick, “Mechanical contribution of lamellar and interla- [39] J. M. Li, A. M. Mullen, S. Yun et al., “Essential role of the phox mellar elastin along the mouse aorta,” Journal of Biomechan- NADPH oxidase subunit p47 in endothelial cell superox- ics, vol. 48, no. 13, pp. 3599–3605, 2015. ide production in response to phorbol ester and tumor necro- [54] R. J. Pelham and Y.-L. Wang, “Cell locomotion and focal sis factor-α,” Circulation Research, vol. 90, no. 2, pp. 143–150, adhesions are regulated by substrate flexibility,” Proceedings of the National Academy of Sciences of the United States of [40] H. Lum and K. A. Roebuck, “Oxidant stress and endothelial America, vol. 94, no. 25, pp. 13661–13665, 1997. cell dysfunction,” American Journal of Physiology-Cell [55] J. P. Califano and C. A. Reinhart-King, “A balance of sub- Physiology, vol. 280, no. 4, pp. C719–C741, 2001. strate mechanics and matrix chemistry regulates endothelial [41] S. van Wetering, J. D. van Buul, S. Quik et al., “Reactive cell network assembly,” Cellular and Molecular Bioengineer- oxygen species mediate Rac-induced loss of cell-cell adhesion ing, vol. 1, no. 2-3, pp. 122–132, 2008. in primary human endothelial cells,” Journal of Cell Science, [56] A. Engler, L. Bacakova, C. Newman, A. Hategan, M. Griffin, vol. 115, no. 9, pp. 1837–1846, 2002. and D. Discher, “Substrate compliance versus ligand density [42] P. V. Usatyuk and V. Natarajan, “Regulation of reactive oxy- in cell on gel responses,” Biophysical Journal, vol. 86, no. 1, gen species-induced endothelial cell-cell and cell-matrix con- pp. 617–628, 2004. tacts by focal adhesion kinase and adherens junction [57] A. J. Engler, C. Carag-Krieger, C. P. Johnson et al., “Embry- proteins,” American Journal of Physiology-Lung Cellular and onic cardiomyocytes beat best on a matrix with heart-like Molecular Physiology, vol. 289, no. 6, pp. L999–L1010, elasticity: scar-like rigidity inhibits beating,” Journal of Cell Science, vol. 121, no. 22, pp. 3794–3802, 2008. [43] C. G. Kevil, N. Ohno, D. C. Gute et al., “Role of cadherin [58] P. Moshayedi, L. da F Costa, A. Christ et al., “Mechanosensi- internalization in hydrogen peroxide-mediated endothelial tivity of astrocytes on optimized polyacrylamide gels ana- permeability,” Free Radical Biology and Medicine, vol. 24, lyzed by quantitative morphometry,” Journal of Physics: no. 6, pp. 1015–1022, 1998. Condensed Matter, vol. 22, no. 19, p. 194114, 2010. [44] G. R. Drummond, S. Selemidis, K. K. Griendling, and C. G. [59] P. W. Oakes, D. C. Patel, N. A. Morin et al., “Neutrophil mor- Sobey, “Combating oxidative stress in vascular disease: phology and migration are affected by substrate elasticity,” NADPH oxidases as therapeutic targets,” Nature Reviews Blood, vol. 114, no. 7, pp. 1387–1395, 2009. Drug Discovery, vol. 10, no. 6, pp. 453–471, 2011. [60] S. R. Peyton and A. J. Putnam, “Extracellular matrix rigidity [45] N. R. Madamanchi, A. Vendrov, and M. S. Runge, “Oxidative governs smooth muscle cell motility in a biphasic fashion,” stress and vascular disease,” Arteriosclerosis Thrombosis and Journal of Cellular Physiology, vol. 204, no. 1, pp. 198–209, Vascular Biology, vol. 25, no. 1, pp. 29–38, 2005. [46] J. E. Wagenseil, N. L. Nerurkar, R. H. Knutsen, R. J. Okamoto, [61] A. J. Engler, M. A. Griffin, S. Sen, C. G. Bönnemann, H. L. D. Y. Li, and R. P. Mecham, “Effects of elastin haploinsuffi- Sweeney, and D. E. Discher, “Myotubes differentiate opti- ciency on the mechanical behavior of mouse arteries,” Amer- mally on substrates with tissue-like stiffness,” The Journal of ican Journal of Physiology-Heart and Circulatory Physiology, Cell Biology, vol. 166, no. 6, pp. 877–887, 2004. vol. 289, no. 3, pp. H1209–H1217, 2005. [62] A. J. Engler, S. Sen, H. L. Sweeney, and D. E. Discher, “Matrix [47] B. A. Sather, D. Hageman, and J. E. Wagenseil, “Murray’s law elasticity directs stem cell lineage specification,” Cell, vol. 126, in elastin haploinsufficient (Eln+/-) and wild-type (WT) no. 4, pp. 677–689, 2006. mice,” Journal of Biomechanical Engineering, vol. 134, no. 12, p. 124504, 2012. [63] T. A. Ulrich, E. M. de Juan Pardo, and S. Kumar, “The mechanical rigidity of the extracellular matrix regulates the [48] A. Hilgendorff, K. Parai, R. Ertsey et al., “Lung matrix and structure, motility, and proliferation of glioma cells,” Cancer vascular remodeling in mechanically ventilated elastin Research, vol. 69, no. 10, pp. 4167–4174, 2009. haploinsufficient newborn mice,” American Journal of Physiology-Lung Cellular and Molecular Physiology, vol. 308, [64] S. R. Peyton, C. B. Raub, V. P. Keschrumrus, and A. J. no. 5, pp. L464–L478, 2015. Putnam, “The use of poly(ethylene glycol) hydrogels to Applied Bionics and Biomechanics 13 [79] R. P. Brandes, N. Weissmann, and K. Schröder, “Nox family investigate the impact of ECM chemistry and mechanics on smooth muscle cells,” Biomaterials, vol. 27, no. 28, NADPH oxidases in mechano-transduction: mechanisms pp. 4881–4893, 2006. and consequences,” Antioxidants & Redox Signaling, vol. 20, no. 6, pp. 887–898, 2014. [65] J. A. Wood, N. M. Shah, C. T. McKee et al., “The role of sub- stratum compliance of hydrogels on vascular endothelial cell [80] J. S. McNally, M. E. Davis, D. P. Giddens et al., “Role of behavior,” Biomaterials, vol. 32, no. 22, pp. 5056–5064, 2011. xanthine oxidoreductase and NAD(P)H oxidase in endothe- lial superoxide production in response to oscillatory shear [66] J. P. Califano and C. A. Reinhart-King, “Substrate stiffness stress,” American Journal of Physiology-Heart and Circula- and cell area predict cellular traction stresses in single cells tory Physiology, vol. 285, no. 6, pp. H2290–H2297, 2003. and cells in contact,” Cellular and Molecular Bioengineering, vol. 3, no. 1, pp. 68–75, 2010. [81] E. Tzima, M. del Pozo, W. B. Kiosses et al., “Activation of Rac1 by shear stress in endothelial cells mediates both cyto- [67] F. J. Byfield, R. K. Reen, T. P. Shentu, I. Levitan, and K. J. skeletal reorganization and effects on gene expression,” The Gooch, “Endothelial actin and cell stiffness is modulated by EMBO Journal, vol. 21, no. 24, pp. 6791–6800, 2002. substrate stiffness in 2D and 3D,” Journal of Biomechanics, vol. 42, no. 8, pp. 1114–1119, 2009. [82] L. H. Yeh, Y. J. Park, R. J. Hansalia et al., “Shear-induced tyrosine phosphorylation in endothelial cells requires Rac1- [68] V. Maruthamuthu, B. Sabass, U. S. Schwarz, and M. L. dependent production of ROS,” American Journal of Gardel, “Cell-ECM traction force modulates endogenous Physiology-Cell Physiology, vol. 276, no. 4, pp. C838–C847, tension at cell-cell contacts,” Proceedings of the National Academy of Sciences of the United States of America, vol. 108, no. 12, pp. 4708–4713, 2011. [83] K. Lee, Q. K. Chen, C. Lui, M. A. Cichon, D. C. Radisky, and C. M. Nelson, “Matrix compliance regulates Rac1b [69] Y. T. Yeh, S. S. Hur, J. Chang et al., “Matrix stiffness regulates localization, NADPH oxidase assembly, and epithelial- endothelial cell proliferation through septin 9,” PLoS One, mesenchymal transition,” Molecular Biology of the Cell, vol. 7, no. 10, 2012. vol. 23, no. 20, pp. 4097–4108, 2012. [70] T. Yeung, P. C. Georges, L. A. Flanagan et al., “Effects of [84] M. A. del Pozo, N. B. Alderson, W. B. Kiosses, H. H. Chiang, substrate stiffness on cell morphology, cytoskeletal structure, R. G. Anderson, and M. A. Schwartz, “Integrins regulate Rac and adhesion,” Cell Motility and the Cytoskeleton, vol. 60, targeting by internalization of membrane domains,” Science, no. 1, pp. 24–34, 2005. vol. 303, no. 5659, pp. 839–842, 2004. [71] K. M. Stroka and H. Aranda-Espinoza, “Endothelial cell [85] B. Wojciak-Stothard and A. J. Ridley, “Rho GTPases and the substrate stiffness influences neutrophil transmigration via regulation of endothelial permeability,” Vascular Pharmacol- myosin light chain kinase-dependent cell contraction,” Blood, ogy, vol. 39, no. 4-5, pp. 187–199, 2002. vol. 118, no. 6, pp. 1632–1640, 2011. [86] J. M. Carbajal, M. L. Gratrix, C. H. Yu, and R. C. Schaeffer Jr., [72] I. Dalle-Donne, R. Rossi, A. Milzani, P. di Simplicio, and “ROCK mediates thrombin’s endothelial barrier dysfunc- R. Colombo, “The actin cytoskeleton response to oxidants: tion,” American Journal of Physiology-Cell Physiology, from small heat shock protein phosphorylation to changes vol. 279, no. 1, pp. C195–C204, 2000. in the redox state of actin itself,” Free Radical Biology and Medicine, vol. 31, no. 12, pp. 1624–1632, 2001. [87] A. A. Birukova, K. Smurova, K. G. Birukov, K. Kaibuchi, J. G. N. Garcia, and A. D. Verin, “Role of Rho GTPases in [73] Q. Xu, L. P. Huff, M. Fujii, and K. K. Griendling, “Redox thrombin-induced lung vascular endothelial cells barrier dys- regulation of the actin cytoskeleton and its role in the vascular function,” Microvascular Research, vol. 67, no. 1, pp. 64–77, system,” Free Radical Biology and Medicine, vol. 109, pp. 84–107, 2017. [88] N. Taulet, V. D. Delorme-Walker, and C. DerMardirossian, [74] K. Buyukhatipoglu and A. M. Clyne, “Superparamagnetic “Reactive oxygen species regulate protrusion efficiency by iron oxide nanoparticles change endothelial cell morphology controlling actin dynamics,” PLoS One, vol. 7, no. 8, 2012. and mechanics via reactive oxygen species formation,” Jour- nal of Biomedical Materials Research Part A, vol. 96A, no. 1, [89] S. Esser, M. G. Lampugnani, M. Corada, E. Dejana, and pp. 186–195, 2011. W. Risau, “Vascular endothelial growth factor induces VE-cadherin tyrosine phosphorylation in endothelial cells,” [75] L. E. Crawford, E. E. Milliken, K. Irani et al., “Superoxide Journal of Cell Science, vol. 111, pp. 1853–1865, 1998. mediated actin response in post-hypoxic endothelial cells,” Journal of Biological Chemistry, vol. 271, no. 43, pp. 26863– [90] P. Andriopoulou, P. Navarro, A. Zanetti, M. G. Lampugnani, 26867, 1996. and E. Dejana, “Histamine induces tyrosine phosphorylation of endothelial cell-to-cell adherens junctions,” Arteriosclero- [76] L. Moldovan, N. I. Moldovan, R. H. Sohn, S. A. Parikh, and sis, Thrombosis, and Vascular Biology, vol. 19, no. 10, P. J. Goldschmidt-Clermont, “Redox changes of cultured pp. 2286–2297, 1999. endothelial cells and actin dynamics,” Circulation Research, vol. 86, no. 5, pp. 549–557, 2000. [91] F. E. Nwariaku, Z. Liu, X. Zhu et al., “NADPH oxidase medi- ates vascular endothelial cadherin phosphorylation and [77] A. Aghajanian, E. S. Wittchen, S. L. Campbell, and endothelial dysfunction,” Blood, vol. 104, no. 10, pp. 3214– K. Burridge, “Direct activation of RhoA by reactive oxygen 3220, 2004. species requires a redox-sensitive motif,” PLoS One, vol. 4, no. 11, article e8045, 2009. [92] R. E. Huie and S. Padmaja, “The reaction of no with superox- ide,” Free Radical Research Communications, vol. 18, no. 4, [78] K. M. Mohazzab, P. M. Kaminski, and M. S. Wolin, “NADH pp. 195–199, 1993. oxidoreductase is a major source of superoxide anion in bovine coronary artery endothelium,” American Journal of [93] S. Moncada, R. M. Palmer, and E. A. Higgs, “Nitric-oxide - Physiology-Heart and Circulatory Physiology, vol. 266, no. 6, physiology, pathophysiology, and pharmacology,” Pharma- pp. H2568–H2572, 1994. cological Reviews, vol. 43, no. 2, pp. 109–142, 1991. 14 Applied Bionics and Biomechanics [94] T. Walsh, T. Donnelly, and D. Lyons, “Impaired endothelial nitric oxide bioavailability: a common link between aging, hypertension, and atherogenesis?,” Journal of the American Geriatrics Society, vol. 57, no. 1, pp. 140–145, 2009. [95] U. Landmesser, S. Dikalov, S. R. Price et al., “Oxidation of tetrahydrobiopterin leads to uncoupling of endothelial cell nitric oxide synthase in hypertension,” Journal of Clinical Investigation, vol. 111, no. 8, pp. 1201–1209, 2003. [96] F. Cosentino, J. E. Barker, M. P. Brand et al., “Reactive oxygen species mediate endothelium-dependent relaxations in tetrahydrobiopterin-deficient mice,” Arteriosclerosis, Throm- bosis, and Vascular Biology, vol. 21, no. 4, pp. 496–502, [97] D. Stuehr, S. Pou, and G. M. Rosen, “Oxygen reduction by nitric-oxide synthases,” Journal of Biological Chemistry, vol. 276, no. 18, pp. 14533–14536, 2001. [98] I. V. Turko and F. Murad, “Protein nitration in cardiovascu- lar diseases,” Pharmacological Reviews, vol. 54, no. 4, pp. 619– 634, 2002. [99] C. A. Reinhart-King, “How matrix properties control the self- assembly and maintenance of tissues,” Annals of Biomedical Engineering, vol. 39, no. 7, pp. 1849–1856, 2011. [100] D. E. Discher, P. Janmey, and Y. L. Wang, “Tissue cells feel and respond to the stiffness of their substrate,” Science, vol. 310, no. 5751, pp. 1139–1143, 2005. [101] K. Roy, Y. Wu, J. L. Meitzler et al., “NADPH oxidases and cancer,” Clinical Science, vol. 128, no. 12, pp. 863–875, 2015. [102] V. Sosa, T. Moliné, R. Somoza, R. Paciucci, H. Kondoh, and M. E. LLeonart, “Oxidative stress and cancer: an overview,” Ageing Research Reviews, vol. 12, no. 1, pp. 376–390, 2013. [103] J. C. Kohn, D. W. Zhou, F. Bordeleau et al., “Cooperative effects of matrix stiffness and fluid shear stress on endothelial cell behavior,” Biophysical Journal, vol. 108, no. 3, pp. 471– 478, 2015. [104] P. A. Galie, A. van Oosten, C. S. Chen, and P. A. Janmey, “Application of multiple levels of fluid shear stress to endo- thelial cells plated on polyacrylamide gels,” Lab on a Chip, vol. 15, no. 4, pp. 1205–1212, 2015. [105] W. O. Carter, P. K. Narayanan, and J. P. Robinson, “Intracel- lular hydrogen peroxide and superoxide anion detection in endothelial cells,” Journal of Leukocyte Biology, vol. 55, no. 2, pp. 253–258, 1994. International Journal of Advances in Rotating Machinery Multimedia Journal of The Scientific Journal of Engineering World Journal Sensors Hindawi Hindawi Publishing Corporation Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 http://www www.hindawi.com .hindawi.com V Volume 2018 olume 2013 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Journal of Control Science and Engineering Advances in Civil Engineering Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Submit your manuscripts at www.hindawi.com Journal of Journal of Electrical and Computer Robotics Engineering Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 VLSI Design Advances in OptoElectronics International Journal of Modelling & Aerospace International Journal of Simulation Navigation and in Engineering Engineering Observation Hindawi Hindawi Hindawi Hindawi Volume 2018 Volume 2018 Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com www.hindawi.com www.hindawi.com Volume 2018 International Journal of Active and Passive International Journal of Antennas and Advances in Chemical Engineering Propagation Electronic Components Shock and Vibration Acoustics and Vibration Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018

Journal

Applied Bionics and BiomechanicsHindawi Publishing Corporation

Published: Apr 14, 2019

References