Advanced age protects resistance arteries of mouse skeletal muscle from oxidative stress through attenuating apoptosis induced by hydrogen peroxide

Abstract

Advanced age is associated with elevated oxidative stress and can protect the endothelium from cell death induced by H₂O₂. Whether such protection occurs for intact vessels or differs between smooth muscle cell (SMC) and endothelial cell (EC) layers is unknown. We tested the hypothesis that ageing protects SMCs and ECs during acute exposure to H₂O₂ (200 µM, 50 min). Mouse superior epigastric arteries (SEAs; diameter, ~150 μm) were isolated and pressurized to 100 cm H₂O at 37˚C. For SEAs from young (4 mo) mice, H₂O₂ killed 57% of SMCs and 11% of ECs in males vs. 8% and 2%, respectively, in females. Therefore, SEAs from males were studied to resolve the effect of ageing and experimental interventions. For old (24 mo) mice, SMC death was reduced to 10% with diminished accumulation of [Ca²⁺]_i in the vessel wall during H₂O₂ exposure. In young mice, genetic deletion of IL-10 mimicked the protective effect of ageing on cell death and [Ca²⁺]_i accumulation. Whereas endothelial denudation or gap junction inhibition (carbenoxolone; 100 μM) increased SMC death, inhibiting NO synthase (L-NAME, 100 μM) or scavenging peroxynitrite (FeTPPS, 5 μM) reduced SMC death along with [Ca²⁺]_i. Despite NO toxicity via peroxynitrite formation, endothelial integrity protects SMCs. Caspase inhibition (Z-VAD-FMK, 50 μM) attenuated cell death with immunostaining for annexin V, cytochrome C, and caspases 3 and 9 pointing to induction of intrinsic apoptosis during H₂O₂ exposure. We conclude that advanced age reduces Ca²⁺ influx that triggers apoptosis, thereby promoting resilience of the vascular wall during oxidative stress.

INTRODUCTION

Vascular levels of reactive oxygen species (ROS) increase with advanced age (Muller-Delp et al., 2012;Moon et al., 2001). Acute exposure to H₂O₂ leads to increased intracellular Ca²⁺ concentration ([Ca²⁺]_i) that results in endothelial cell (EC) (Sun et al., 2012;Socha et al., 2015) and smooth muscle cell (SMC) death (Li et al., 2003;Kalyankrishna et al., 2002;Jin et al., 2000). For endothelial tubes freshly isolated from mouse superior epigastric arteries (SEAs), Ca²⁺ entry and EC death induced by H₂O₂ (200 μM) were attenuated by ~80% in aged (24 months) vs. young (4 months) mice (Socha et al., 2015). Whereas adaptation to chronic oxidative stress may explain EC resilience to H₂O₂ during advanced age, it is unknown whether similar adaptation occurs in SMCs, nor whether respective cell layers protect each other in the intact vessel wall. In this study, we tested the hypothesis that advanced age protects intact SEAs from H₂O₂-induced cell death by restricting aberrant elevation of [Ca²⁺]_i.

Interleulkin-10 knockout mice (IL-10^−/−) also manifest elevated levels of vascular oxidative stress (Kinzenbaw et al., 2013;Sikka et al., 2013) and mimic the phenotype of frailty during advanced age (Walston et al., 2008;Ko et al., 2011). Therefore, to independently test whether chronic oxidative stress confers resilience to the adverse effects of H₂O₂ on the vascular wall, we performed complementary experiments on SEAs from young IL-10^−/− mice. Our findings illustrate that for SEAs of young mice acutely exposed to H₂O₂ (200 μM for 50 min), SMCs are more susceptible than ECs to apoptosis induced by Ca²⁺ overload, with vessels from females more resilient than vessels from males. Remarkably, both advanced age and genetic deletion of IL-10 in young mice protected SMCs from apoptosis in association with a reduction in vessel wall [Ca²⁺]_i. In contrast to nitric oxide (NO) being cytoprotective (Polte et al., 1997), it was found to be cytotoxic through the generation of peroxynitrite. Denuding the endothelium increased [Ca²⁺]_i and cell death, indicating that integrity of the endothelium promotes vascular resilience during acute oxidative stress.

METHODS

Ethical Approval

All procedures were approved by the Animal Care and Use Committee of the University of Missouri, Columbia (approval reference no. ACUC #9220), were performed in accord with the National Research Council’s Guide for the Care and Use of Laboratory Animals (2011), and in compliance with the animal ethics checklist of this journal.

Animal Care and Use

Experiments were performed on young (n=115, 3–4 months old; purchased from Jackson Laboratories: Bar Harbor, ME, USA) and old (n=28, 24–26 months old; National Institute on Ageing, NIA) C57BL/6J mice. Our initial experiments found SMCs and ECs in SEAs from young males to be more susceptible to H₂O₂ compared to those in age-matched females (see Figure 1). Therefore, SEAs from male mice were studied to resolve the effect of experimental manipulations. Complementary experiments were performed on male IL-10^−/− mice bred on a C57BL/6J background (n=14; ~4 months old, Jackson stock #002251, B6.129P2-Il10^tm1Cgn/J). All mice were housed on a 12:12 h light-dark cycle at ∼23°C with fresh water and food available ad libitum. Each mouse was anaesthetized with ketamine + xylazine (100 mg kg⁻¹ and 10 mg kg⁻¹ respectively; intraperitoneal injection) to harvest tissues and then killed by exsanguination. The order in which mice from respective groups were studied was randomized.

Figure 1. H₂O₂-induced death is greater in SMCs vs. ECs and in males vs. females.

A, PI staining nuclei of dead cells following 50 min in control PSS, 50 min of H₂O₂ exposure or endothelial denudation in control PSS. B, Hoechst 33342 dye staining nuclei of all cells in respective panels of A. C, Merged images of panels in A and B. Images acquired from pressurized (100 cm H₂O) SEAs from young male mice. There was an average of 115 SMCs and 82 ECs in SEAs from males and 120 SMCs and 81 ECs in SEAs from females in each ROI (dotted rectangle). D, Number of live and dead ECs and SMCs in SEAs from male and female mice following exposure to H₂O₂ (200 μM; 50 min). E, SMC and EC death following H₂O₂ exposure expressed as % of respective total cell numbers in D. F, SMC death (% of total SMC number) following H₂O₂ exposure from intact or endothelial-denuded vessels from male mice. Values are means ± SEM for n = 9 per group for males and n = 4 per group for females. Scale bar in C = 50 µm and applies to all panels. *P < 0.05 vs. SMC. ^τ P < 0.05 female vs. male. ^#P < 0.05 vs. intact. Small circular nuclei are adventitial cells and were not counted.

Experimental preparation of superior epigastric arteries

On the morning of an experiment, abdominal fur was removed from the anaesthetized mouse by shaving and a midline incision through the skin was made from the sternum to the pubis. The abdominal muscles were exposed, removed bilaterally and placed in chilled (4°C), nominally Ca²⁺-free physiological salt solution (PSS; pH 7.4) containing (in mM): 140 NaCl (Fisher Scientific; Pittsburgh, PA, USA), 5 KCl (Fisher), 1 MgCl₂ (Sigma-Aldrich; St. Louis, MO, USA), 10 HEPES (Sigma), and 10 glucose (Fisher). Muscles were pinned as a flat sheet onto transparent silicone rubber (Sylgard 184; Dow Corning; Midland, MI USA) and an unbranched segment of the SEA (length, ∼2 mm; diameter, ~150 µm) was dissected from the surrounding tissue while viewing through a stereomicroscope. During dissection, sodium nitroprusside (1 µM) was added to the PSS to relax SMCs. Following isolation, individual SEAs were transferred to a tissue chamber (RC-37N; Warner Instruments; Hamden, CT, USA) secured in a platform with micromanipulators (MT-XYZ; Siskiyou Corp.; Grants Pass, OR, USA) holding heat-polished cannulation micropipettes (external diameter, ∼100 µm). The SEA was cannulated at both ends, each end was tied onto its cannula with a strand of silk suture, and the preparation was transferred to the stage of a Nikon 600FN microscope (Tokyo, Japan). The vessel was pressurized to 100 cm H₂O (~75 mmHg) and superfused in the tissue chamber at 3 mL min⁻¹ with control PSS containing 2 mM CaCl₂ (Fisher) while maintained at 37°C and allowed to equilibrate for 30 min. Vessel inner diameter (ID) was measured (spatial resolution, 1 μm) from images acquired through a 20X objective (Nikon Fluor20, numerical aperture = 0.45) onto a charge coupled device camera (MyoCam-S; IonOptix; Milford, MA, USA) (Norton & Segal, 2018). Spontaneous resting tone (%) was calculated as: [(ID_max – ID_rest)/ID_max] × 100 where ID_rest = resting diameter following equilibration and ID_max = ID under Ca²⁺ free conditions.

Endothelial denudation.

To evaluate the contribution of ECs to SMC responses, the endothelium was disrupted by perfusing an air bubble through the vessel lumen. Selective loss and disruption of ECs was verified by the staining of any residual EC (but not SMC) nuclei with propidium iodide (PI) along the denuded segment or loss of vasodilation to acetylcholine (Bartlett & Segal, 2000;Emerson & Segal, 2000a) in vessels preconstricted with noradrenaline [NA; 170 nM = EC₅₀ (Boerman & Segal, 2016)]. To confirm that endothelial function remained impaired for the duration of an experimental protocol, responses to acetylcholine were evaluated 3 h after initial assessment of EC disruption.

Cell Death

To quantify cell death, cannulation pipettes were preloaded with PSS containing the nuclear stains Hoechst 33342 (1 µM; Cat. #H1399, Fisher), a membrane-permeant dye that stains all cell nuclei, and PI (2 µM; Cat. #P4170, Sigma), which enters cells upon membrane disruption and stains nuclei as a marker of lethal injury (Bartlett & Segal, 2000;Emerson & Segal, 2000a). Dyes were loaded from within the vessel lumen because bath application led to nonspecific staining in the adventitia that obscured resolution of SMC and EC nuclei. In some experiments, cannulation pipettes also contained annexin V conjugated to Alexa 488 (50 µL mL⁻¹; Cat. #V13241, Fisher), which binds to phosphatidylserine everted in the lipid bilayer of the plasma membrane as a fluorescent marker of apoptosis (Walton et al., 1997). Upon pressurization, the pipette solution entered the lumen of the cannulated vessel with no additional flow through the lumen of a vessel unless otherwise stated.

To induce acute oxidative stress, 200 μM H₂O₂ was added to the superfusion solution for 50 min; PSS lacking H₂O₂ served as the negative control. With reference to our previous studies of endothelial tubes from SEAs (Socha et al., 2015) and extensive preliminary experiments on intact SEAs, these criteria for H₂O₂ exposure were based upon reproducible moderate induction of cell death, which thereby served as a reference for evaluating the effect of experimental interventions. Some preparations were treated with N^G-Nitro-L-arginine methyl ester (L-NAME, 100 µM; Cat. #N5751, Sigma) to inhibit NO synthase; with 5, 10, 15, 20-tetrakis (4-sulfonatophenyl) porphyrinato iron III chloride (FeTPPS, 5 µM; Cat. #341492, EMD Millipore; Burlington, MA, USA) to scavenge peroxynitrite (Stern et al., 1996); or with carbenoxolone (100 µM; Cat. #C4790, Sigma) to inhibit gap junctions (Behringer et al., 2012). Each drug was equilibrated for 20 min prior to, and maintained during, H₂O₂ exposure. Additional experiments examined the role of extracellular Ca²⁺ by replacing control PSS with nominal Ca²⁺-free PSS in the vessel lumen and the superfusion solution; a Ca²⁺ chelator was not included as we found this condition (EGTA, 1 mM) to deplete Ca²⁺ from internal stores and initiate cell death via membrane disruption.

To test the role of caspases in H₂O₂-induced cell death, we used the irreversible membrane-permeant pan-caspase inhibitor carbobenzoxy-valyl-alanyl-aspartyl-[O-methyl]- fluoromethylketone [Z-VAD-FMK, 50 µM; Cat. #ALX-260-020-M001, Enzo Life Sciences, Farmindale, NY, USA (Fazal et al., 2005)] after developing the following protocol: Z-VAD-FMK was added to the tissue chamber for 30 min preincubation without flow at room temperature while the nuclear dyes (PI + Hoechst), H₂O₂ (200 µM), and Z-VAD-FMK were introduced into the vessel lumen via cannulation pipettes. The vessel was then superfused with 200 μM H₂O₂ in control PSS at 37°C. Simultaneously, an intraluminal flow (∼10 µL min⁻¹ = 166 nL s⁻¹) containing 50 μM Z-VAD-FMK and 200 μM H₂O₂ was established with a 5 cm H₂O pressure gradient between inflow and outflow cannulae. Our preliminary experiments established the necessity of maintaining a continuous supply of Z-VAD-FMK within the lumen throughout H₂O₂ exposure and including H₂O₂ in the luminal perfusion solution.

Following H₂O₂ exposure, the vessel chamber and lumen were flushed with control PSS (3 mL min⁻¹ and 100 μL min⁻¹, respectively) for 10 min. Image stacks from the top half a vessel segment were acquired with a 40X water immersion objective [Fluor40; numerical aperture = 0.80] coupled to a DS-Qi2 camera with Elements software (version 4.51) on an E800 microscope (all from Nikon). Images were analyzed in Image J (NIH). Stained nuclei of ECs and SMCs were counted manually within a defined region of interest (ROI; Figure 1). Cell death was calculated as: (# of nuclei stained with PI / total # of nuclei stained with Hoechst 33342) × 100%.

Calcium Photometry

Pressurized intact and denuded SEAs were cannulated as above and placed on an Eclipse TS100 microscope (Nikon). Vessels were equilibrated for ∼20 min at 37°C then incubated for 40 min without flow in Fura-2 AM dye (Cat. #F14185, Invitrogen) added to the bath (final concentration, 1 µM with 0.5% DMSO). Superfusion with control PSS was then resumed to wash out excess dye during 20 min equilibration. We have found this protocol to favor dye uptake into SMCs [(Norton & Segal, 2018) and unpublished observations].

Fura-2 fluorescence was used to monitor [Ca²⁺]_i by alternatively (10 Hz) exciting at 340 and 380 nm while recording emissions at 510 nm through a Fluor20 objective using IonWizard 6.3 software (IonOptix); vessel ID was measured simultaneously from images acquired as described above. After establishing a stable baseline with control PSS, H₂O₂ (200 µM) was included in the superfusion solution. Calcium responses and diameter were measured for 30 s every 5 min (to prevent photo bleaching of Fura-2 dye) during 50 min H₂O₂ exposure and the ensuing 30 min wash with control PSS. For each vessel, Fura-2 signals were obtained from the entire field of view (~300 cells). Values are expressed as F₃₄₀/F₃₈₀ ratios after subtracting autofluorescence recorded prior to dye loading (Norton & Segal, 2018). To determine the contribution of extracellular vs. intracellular Ca²⁺ sources, [Ca²⁺]_i responses to H₂O₂ were measured in the presence of nominally Ca²⁺-free PSS. We evaluated vasoconstriction to NA [170 nM; EC₅₀ (Boerman & Segal, 2016)] in these vessels prior to and following H₂O₂ exposure to evaluate the effect of H₂O₂ exposure on vasomotor function. Reponses were calculated as: [(ID_rest - ID_NA)/ID_rest] × 100, where ID_rest = resting diameter following equilibration and ID_NA = ID in the presence of NA.

Mitochondrial structure and immunostaining for intrinsic apoptosis

Vessels from young male mice (i.e., those most susceptible to cell death; see Figure 1) were superfused with control PSS or with PSS containing H₂O₂ as described above. To examine the effects of H₂O₂ on mitochondrial structure, SEAs were stained with mitotracker deep red (100 nM; Cat. #M22426, Molecular Probes, Eugene OR, USA) (Behringer & Segal, 2017) for 30 min at 37°C in combination with Hoechst 33342 dye (1 µM). For immunostaining, vessels were fixed in 2% paraformaldehyde for 1 h followed by a mild collagenase treatment (1.5 mg mL⁻¹; Cat. #C8051, Sigma) at 37°C for 12–13 min. A SEA was then cut open longitudinally and pinned flat onto silicone elastomer (Sylgard 184; Dow Corning, Midland, MI, USA) using fine tungsten wire (diameter, 25 µm; Cat. #W005130, Goodfellow Corp., Coraopolis, PA, USA) with ECs facing down to facilitate removal of adventitia for imaging SMCs. The vessel was then fixed in 4% paraformaldehyde for 20 min and incubated overnight at 4°C with primary antibodies for cytochrome C (1:100 dilution; Cat. #12963D Lot 2, Cell Signaling Technology, Danvers, MA, USA) (Lee et al., 2016), cleaved caspase 3 (1:50; Cat. #9661S Lot 45, Cell Signaling) (Chand et al., 2018), and cleaved caspase 9 (1:100; Cat. no. 9509S, Lot 3, Cell Signaling) (Monick et al., 2008) in phosphate buffered saline (PBS). After washing the preparations (3 × 5 min) in PBS, an Alexa Fluor 488 secondary antibody (1:200) was incubated for 2 h at room temperature then washed in PBS. Controls for nonspecific staining used only primary or secondary antibodies (Boerman & Segal, 2016). The nuclear dye, TO-PRO-3 iodide (Cat. #T3605, Fisher) was added (2 µM) for 20 min at room temperature. After a final wash, vessel preparations were placed on a glass slide with the ECs facing up, cover-slipped and sealed with ProLong Gold (Cat. #P36930, Fisher) for imaging on a Leica SP8 confocal laser-scanning inverted microscope (Leica Microsystems; Buffalo Grove, IL, USA). Images were acquired using an HC PL APO 63X oil immersion objective (numerical aperture = 1.4; Leica) and 0.5 μm Z-slices. Confocal images are maximal intensity Z-stack projections.

Statistics

Data were analyzed using unpaired Student’s t-tests or Analysis of Variance (Prism 5, GraphPad Software; La Jolla, Ca, USA) as appropriate. When significant main effects were detected with ANOVA, post hoc comparisons were performed using Bonferroni tests. A probability of P<0.05 was accepted as statistically significant. Summary data are presented as means ± SEM where n refers to the number of vessels in an experimental group with 1 SEA studied per mouse. Findings are from male mice unless stated otherwise.

RESULTS

We established 200 µM H₂O₂ as appropriate for experimental protocols < 2 h duration (including 30 min equilibration) based on preliminary findings that [H₂O₂] ≤ 100 µM required several h for its effects to become manifest while [H₂O₂] ≥ 500 µM compromised cell viability within several min. Perfusion of 200 µM H₂O₂ through the vessel lumen with control PSS superfused in the bath resulted in an axial gradient of cell death along the vessel wall, attributable to a progressive decline in luminal [H₂O₂] as it diffused into the bath. In contrast, including 200 µM H₂O₂ in the bath established a constant source of H₂O₂ with a uniform pattern of cell death along the vessel wall that was not different from simultaneously perfusing H₂O₂ through the vessel lumen. Thus, a single layer of SMCs did not impair H₂O₂ diffusion between the bath and endothelium.

H₂O₂-induced death is greater in SMCs vs. ECs and in males vs. females

Nuclear staining identifies ECs (oval nuclei oriented parallel to vessel axis) vs. SMCs (slender nuclei oriented perpendicular to vessel axis; Figures 1A–C), consistent with our previous findings in hamster microvessels (Bartlett & Segal, 2000;Emerson & Segal, 2000b). Following 50 min exposure to H₂O₂, total cells (nuclei stained with Hoechst 33342) and dead cells (nuclei stained with PI) were counted within ROIs encompassing ~100 cells of each type (Figure 1D). Cell death was calculated as: (# of nuclei stained with PI / total # of nuclei stained with Hoechst 33342) and dead cells were expressed relative to total cell counts (Figures 1E and 1F). In SEAs from young male mice, 57% of SMCs and 11% of ECs stained with PI. Consistent with protection from vascular apoptosis in female mice (Spyridopoulos et al., 1997), SEAs from females had 8% of SMCs and 2% of ECs stained with PI following equivalent exposure to H₂O₂ with no difference in the number of either cell type. The several-fold greater effect of H₂O₂ on SMCs and ECs found for SEAs from males is the basis of using males to resolve the effects of ensuing protocols. For SEAs from males, denuding the endothelium nearly doubled SMC death to 92% (Figure 1F). We confirmed that neither SMCs nor ECs stained with PI in the absence of H₂O₂ (≤ 1%; Figures 1A–C, left panels), nor did endothelial denudation result in SMC death for SEAs maintained in control PSS (Figure 1A–C, right panel). Endothelial function remained impaired (dilatation in response to 10 µM acetylcholine = 7 ± 4%; n = 3) for > 3 h following denudation (vs. 93 ± 2% dilatation with intact endothelium; n = 4).

Nitric oxide and gap junctions differentially affect SMC death induced by H₂O₂

Finding that ECs exerted a protective effect on SMCs during exposure to H₂O₂ (Figure 1F), we tested whether NO mediated SMC protection (Polte et al., 1997). Contrary to this hypothesis, inhibition of NO synthase (L-NAME; 100 μM) reduced SMC death by more than half (P<0.05) with no effect on EC viability (Figures 2A and 2B). This outcome suggested that NO (by reacting with superoxide) led to the generation of peroxynitrite, which can damage ECs and SMCs (Dickhout et al., 2005;O’Connor et al., 1997). Accordingly, the peroxynitrite scavenger FeTPPs significantly reduced death of both SMCs and ECs (Figures 2A and 2B). The lack of effect of FeTTPs on SMC death in endothelium denuded vessels (Figure 2C) supports the interpretation that endothelium-derived NO is required for peroxynitrite formation.

Figure 2. Nitric oxide and gap junctions differentially affect SMC death induced by H₂O₂.

Cell death (% PI positive cells) in SMCs (A, C, D, F) and ECs (B, E) from SEAs of young male mice following H₂O₂ exposure (200 μM, 50 min). SMC death was attenuated by inhibiting NO synthase with L-NAME (100 µM) and scavenging peroxynitrite with FeTPPS (5 µM) while the gap junction inhibitor carbenoxolone (CBX, 100 μM) increased SMC death. EC death was several-fold lower than SMC death and further reduced by FeTPPS. Values are means ± SEM for n = 4–9 per group. *P < 0.05 vs. control.

Electrical coupling between SMCs and ECs occurs through myoendothelial gap junctions in resistance arteries similar in structure and function to the SEA (Emerson & Segal, 2000a). Including the gap junction blocker carbenoxolone [100 μΜ (Behringer et al., 2012)] during H₂O₂ exposure increased SMC death by an additional 20% (Figure 2D), suggesting that coupling to ECs exerts a protective effect on SMCs during oxidative stress. In contrast, there was no effect of carbenoxolone on EC death (Figure 2E), which was several-fold lower that SMC death (Figure 2E). In vessels denuded of ECs and exposed to H₂O₂, carbenoxolone did not alter SMC death, which was already near total (Figures 2F). In the absence of H₂O₂, carbenoxolone had a negligible effect on cell death (EC = 5±3%, SMC = 2±1%; n=3).

NO and peroxynitrite augment H₂O₂-induced Ca²⁺ entry

Exposure to H₂O₂ can increase [Ca²⁺]_i (Socha et al., 2015;Sun et al., 2012) and elevated [Ca²⁺]_i can initiate cell death (Kajstura et al., 1997). Therefore, we assessed the effects of L-NAME and FeTPPs on [Ca²⁺]_i responses to H₂O₂ exposure. Consistent with their effects on cell death, inhibiting NO synthesis or scavenging peroxynitrite reduced [Ca²⁺]_i responses during exposure to H₂O₂ (Figures 3A and 3B). To determine whether this rise in [Ca²⁺]_i is attributable to Ca²⁺ influx through the plasma membrane, [Ca²⁺]_i responses were evaluated in arteries with nominally Ca²⁺-free PSS in the bath and lumen. The increase in [Ca²⁺]_i during H₂O₂ was nearly eliminated when 2 mM Ca²⁺ was absent from the PSS (Figure 3C).

Figure 3. NO and peroxynitrite augment H₂O₂-induced Ca²⁺ entry.

A, Baseline Ca²⁺ (F₃₄₀/F₃₈₀ ratios) from intact SEAs of young male mice in control PSS and in the presence of L-NAME or FeTPPs. B, Changes in vessel wall [Ca²⁺]_i from baseline during H₂O₂ exposure (200 μM, 50 min) and 30 min washout with control PSS; inhibiting NO synthesis or scavenging peroxynitrite reduced [Ca²⁺]_i in response to H₂O₂. C, Vessel wall [Ca²⁺]_i responses to H₂O₂ are nearly abolished with Ca²⁺-free PSS. Baseline F₃₄₀/F₃₈₀ ratios were 0.91±0.08 in control PSS containing 2 mM [Ca²⁺] and 0.80±0.08 in nominally Ca²⁺-free PSS. Values are means ± SEM for n = 4–7 per group. *P < 0.05 vs. control PSS.

Advanced age promotes cell survival during acute oxidative stress

In endothelial tubes freshly isolated from SEAs and exposed to 200 μM H₂O₂, advanced age reduced EC death (Socha et al., 2015). However the effect of ageing on the resilience of SMCs to a similar stress was unknown. As shown in Figure 4A, SMC death was several-fold lower for intact SEAs from old mice compared to young mice. While endothelial denudation increased SMC death, the effect of advanced age remained. Consistent with our original findings in endothelial tubes (Socha et al., 2015), H₂O₂-induced death of ECs in intact vessels was also lower with advanced age (Figure 4B).

Figure 4. Advanced age promotes cell survival during acute oxidative stress.

A, SMC death in response to H₂O₂ exposure (200 μM; 50 min) for intact and denuded SEAs from young and old male mice. B, EC death for intact SEAs from young and old mice in response to H₂O₂. Vessels from old mice consistently had lower cell death vs. vessels from young mice. The total number of SMC nuclei was not different between groups (young = 115 ± 2; old = 111 ± 3). There was a greater number of EC nuclei in old (93± 2) vs. young (82 ± 3) SEAs. Values are means ± SEM for n = 9–11 per group. *P < 0.05 vs. Young. ^#P < 0.05 vs. Intact.

Advanced age protects SMCs from H₂O₂-induced Ca²⁺ entry

Under resting conditions in control PSS, baseline [Ca²⁺]_i was elevated in denuded SEAs from old mice (Figure 5A) relative to other groups. Exposure to H₂O₂ increased [Ca²⁺]_i through 50 min (Figure 5B) which reversed completely during 30 min washout for intact SEAs from old but not young mice. Consistent with differences in cell death between age groups, the rise in [Ca²⁺]_i for intact SEAs from young mice was three-fold greater than for SEAs from old mice. While there was no significant difference in the Ca²⁺ response to H₂O₂ between intact and denuded SEAs from young mice, denudation augmented the Ca²⁺ response of SMCs in vessels from old mice several-fold (Figure 5B). Time controls in the absence of H₂O₂ confirmed that [Ca²⁺]_i (Figure 5C) and vessel diameter (Figure 5D) remained stable (and not different) for both age groups throughout the duration of our protocols.

Figure 5. Advanced age protects SMCs from H₂O₂-induced Ca²⁺ entry.

A, Baseline Ca²⁺ (F₃₄₀/F₃₈₀ ratios) from intact and denuded SEAs from young and old mice. B, Changes in vessel wall [Ca²⁺]_i from baseline during H₂O₂ (200 μM, 50 min) and 30 min washout with control PSS. [Ca²⁺]_i was reduced in old vs. young. C and D, Stability of Ca²⁺]_i and diameter of intact SEAs for the duration of experimental protocols; initial datum in D indicates maximum (Max) diameter in Ca²⁺ free PSS. E, SEA diameter during 50 min H₂O₂ exposure (note transient constriction) and 30 min washout with control PSS; data points preceding 0 min indicates Ca²⁺-free [maximum (Max)] diameter. F and G, Vasoconstriction to NA (170 nM) pre- and post-H₂O₂ exposure, respectively, for intact and denuded SEAs; note greater retention in SEAs from old vs. young mice following H₂O₂. H, Stability of time controls for vasoconstriction to NA prior to and following 80 min incubation in control PSS. Values are means ± SEM for n = 4–7 per group. *P < 0.05 vs. young. ^#P < 0.05 vs. intact. ^§P < 0.05 vs. pre-H₂O₂ exposure. YI, young intact; OI, old intact; YD, young denuded; OD, old denuded.

Spontaneous resting tone was ~10% and exposure to H₂O₂ transiently constricted vessels by another ~20% in all groups (Figure 5E); vessel diameter remained stable thereafter while [Ca²⁺]_i continued to rise (Figure 5B). Prior to H₂O₂ exposure, addition of NA (170 nM; EC₅₀) elicited 40–50% constriction in vessels from each group (Figure 5F). Following H₂O₂ exposure, vasoconstriction was diminished (Figure 5G) in accordance with reduced SMC viability. This loss of adrenergic constriction was greater for intact SEAs from young vs. old mice and was further reduced by endothelial denudation in both age groups. Time controls confirmed that vasoconstriction to NA for intact SEAs was maintained for the duration of experiments when vessels remained in control PSS (Figure 5H).

Absence of IL-10 attenuates H₂O₂-induced cell death and Ca²⁺ influx

Loss of IL-10 results in chronic inflammation, EC oxidative stress and vascular dysfunction (Kinzenbaw et al., 2013;Sikka et al., 2013) with IL-10^−/− mice used as a model of advanced age (Walston et al., 2008;Ko et al., 2011). As seen for SEAs from old mice, SMC and EC death was reduced in SEAs of young IL-10^−/− mice vs. young controls (Figures 6A and 6B). With no difference in baseline [Ca²⁺]_i (Figure 6C), the [Ca²⁺]_i response to H₂O₂ in SEAs from IL-10^−/− mice was less than half of that observed for SEAs from control mice (Figure 6D). Similar to the effects of advanced age (Figure 5G), SEAs from IL-10^−/− mice had greater retention of adrenergic vasoconstriction following H₂O₂ exposure vs. SEAs from control mice (Figure 6F).

Figure 6. Absence of IL-10 attenuates H₂O₂-induced cell death and Ca²⁺ entry.

A, SMC death and B, EC death to H₂O₂ (200 μM; 50 min) were lower for intact SEAs from IL-10^−/− mice vs. control. There was no difference in the total number of SMC or EC nuclei between groups. C, Baseline [Ca²⁺]_i (F₃₄₀/F₃₈₀) was not different between vessels from control and IL-10^−/− mice. D, Changes in vessel wall [Ca²⁺]_i during H₂O₂ (200 μM, 50 min) and 30 min washout with control PSS; [Ca²⁺]_i remained lower in SEAs from IL-10^−/− vs. control mice. E, SEA diameters during H₂O₂ and washout with control PSS. Data points preceding 0 min indicate Ca²⁺-free [maximum (Max)] diameter. F, Post-H₂O₂ exposure, intact SEAs from IL-10^−/− mice retain more noradrenergic vasoconstriction (170 nM NA) vs. Pre H₂O₂ exposure. Values are means ± SEM for n = 4–9 per group. *P < 0.05 vs. control.

Vascular cell death induced by H₂O₂ occurs through intrinsic apoptosis

The modest number of ECs positive for annexin V (Figure 7A), a marker of apoptosis, also exhibited nuclear staining with PI (Figure 7B). Under the same conditions, we were unable to resolve annexin V staining of SMCs, which we attribute to it being retained within the lumen due to its size (MW ~36 kDa). Abluminal application of annexin V resulted in staining throughout the adventitia which precluded reliable imaging of SMCs or ECs in the vessel wall. As an alternate strategy, we used Z-VAD-FMK (50 µM), a pan-caspase inhibitor (Fazal et al., 2005), to test whether vascular cell death was mediated through apoptosis and adjusted our protocol for H₂O₂ exposure as follows. With luminal perfusion of H₂O₂ (and control PSS in the bath), ~20% of SMCs stained with PI (Figure 7C). With H₂O₂ in the bath and the luminal perfusion solution, SMC death increased to ~50%; under these conditions, preincubating with Z-VAD-FMK and including it in the luminal perfusion solution reduced SMC death to ~10%. Remarkably, EC death was negligible under all conditions with luminal perfusion (Figure 7D). This outcome contrasts with ~10% EC death in the absence of luminal perfusion (Figures 1 and 2). Further, SMC death during bath and luminal perfusion with H₂O₂ was ~10% less than observed with H₂O₂ in the bath in the absence of luminal perfusion (Figures 1E, 2A, and 4A).

Figure 7. Cell death induced by H₂O₂ occurs through apoptosis.

A, Annexin V staining (green) of ECs illustrates apoptosis in response to H₂O₂ (200 μM, 50 min) in SEAs from young mice. B, Image of annexin V in A merged with PI and Hoechst staining in same vessel; annexin V and PI staining coincide (note 3 red nuclei within respective green ECs). Images are representative of n = 9 vessels. Scale bar in B = 50 µm and applies to A. Cell death (%) in C, SMCs and D, ECs from SEAs exposed to control PSS in the bath solution and in the luminal perfusion solution (“Perfusion Control”); with H₂O₂ only in the lumen; with H₂O₂ in both the lumen and bath; with H₂O₂ in the lumen and bath in the presence of the caspase inhibitor Z-VAD-FMK (Z-VAD, 50 µM). Values are means ± SEM for n = 4–7 per group. *P < 0.05 vs. perfusion control. ^#P < 0.05 vs. H₂O₂ in lumen only. ^§P < 0.05 vs. H₂O₂ in bath and lumen.

Apoptosis through the intrinsic pathway involves cytochrome C release from the mitochondria and the subsequent activation of caspases 9 and 3, respectively (Singh et al., 2007;Viola et al., 2007;Lin et al., 2007;Kroemer et al., 2007;Danial & Korsmeyer, 2004). We focused on quantifying respective events in SMCs due to their greater susceptibility to death compared to ECs during oxidative stress (Figures 1, 2 and 4). Under control conditions, cytochrome C appeared localized within SMCs in accord with mitochondrial compartments (Figures 8A and 8B). Following H₂O₂ exposure, localized staining of cytochrome C was absent from the majority of SMCs. Under control conditions, staining for cleaved (i.e., activated) caspases 9 and 3 was weak (Figures 8C and 8D, respectively). Following H₂O₂ exposure, staining for both caspases was robust in multiple SMCs, with caspase 9 appearing more localized than caspase 3. Control experiments verified lack of nonspecific staining by the secondary antibody (Figures 8E–8H).

Figure 8. Cell death induced by H₂O₂ is mediated through intrinsic apoptosis.

Staining SMCs of intact SEAs for A, mitotracker (red), B, cytochrome C (green; note loss of localization following H₂O₂), C, cleaved caspase 9 (green), and D, cleaved caspase 3 (green); note increased staining following H₂O₂ for both cleaved caspases. Control SEAs maintained in PSS shown in left panels; SEAs exposed to H₂O₂ (200 μM, 50 min) shown in right panels. Cell nuclei are labeled with A, Hoechst 33342 (blue) or B-D, TO-PRO-3 (red). Scale bar in D = 20 µm and applies to all panels in A through D. Images are representative from multiple regions of n = 3 SEAs per group. In the absence of secondary antibody (green), immunofluorescence was negligible for primary antibodies against E, cytochrome C; F, cleaved caspase 9; and G, cleaved caspase 3. H, Secondary antibody fluorescence in the absence of primary antibodies was also negligible. In E through H, SMC nuclei are stained with TO-PRO-3 (red); scale bar in H = 20 µm and applies to panels E–H.

DISCUSSION

We evaluated the resilience of SMCs and ECs of resistance arteries to acute oxidative stress induced by exposure to H₂O₂ (200 μM) for 50 min. For males, SEAs from young mice (4 mo) exhibited 8-fold greater SMC death and 4-fold greater increase in [Ca²⁺]_i compared to those from old (24 mo) mice. When compared to ECs, SMCs were several-fold more susceptible to death induced by H₂O₂ and denuding the endothelium increased SMC death irrespective of age. In vessels from young males, inhibiting NO synthase reduced SMC death, as did scavenging peroxynitrite, its downstream intermediate. In contrast, the inhibition of gap junctions increased SMC death induced by H₂O₂. Genetic deletion of the anti-inflammatory cytokine IL-10 increased the resilience of SMCs to H₂O₂ in the manner observed for advanced age. Diminished cell death in SEAs from old mice and IL-10^−/− mice was associated with attenuated Ca²⁺ entry during H₂O₂ exposure. Immunostaining for markers of apoptosis revealed activation of caspase 9 via cytochrome C release from mitochondria, consistent with intracellular Ca²⁺ overload, and cell death was reduced by inhibiting caspase activity. Resistance arteries from young female mice exhibited resilience to H₂O₂ similar to that observed for old males or those lacking IL-10, serving as in internal control and consistent with the anti-oxidant effects of estrogen (Dantas et al., 2002;Spyridopoulos et al., 1997). These findings are the first to indicate greater susceptibility to oxidative stress in SMCs vs. ECs of intact blood vessels and suggest that advanced age protects vascular cells during oxidative stress by reducing Ca²⁺ influx, an effect that is recapitulated by loss of IL-10. The present data further illustrate that maintaining integrity of the endothelium helps to protect SMCs from the deleterious effects of H₂O₂.

H₂O₂-induced cell death in SMCs: distinct roles of ECs

Our findings illustrate that SMCs are more several-fold more susceptible to damage from H₂O₂ when compared to their adjacent ECs. Furthermore, the resilience of the endothelium extends to protect SMCs, as denudation increased SMC death from H₂O₂. Whereas NO can protect ECs from death via the cGMP pathway through S-nitrosylation of caspases (Grosser & Schroder, 2003;Polte et al., 1997;Li et al., 1997), we found that inhibiting NO synthesis reduced the death of SMCs with no additional effect on ECs. During oxidative stress, NO can react with superoxide to form peroxynitrite (Beckman et al., 1990), which can be lethal to ECs and SMCs (O’Connor et al., 1997;Dickhout et al., 2005). Consistent with this effect, scavenging peroxynitrite with FeTPPS reduced SMC and EC death of intact vessels but not SMCs of vessels in which the endothelium had been denuded. While peroxynitrite may diminish the ability of cells to cope with oxidative stress through inhibiting superoxide dismutase 2 via nitration of tyrosine residues (MacMillan-Crow et al., 1996), peroxynitrite can increase [Ca²⁺]_i via both internal and external sources (Pan et al., 2004). The present findings illustrate that inhibiting NO synthesis or scavenging peroxynitrite greatly attenuated the rise in [Ca²⁺]_i evoked by H₂O₂ (Figure 3).

When our data from intact SEAs are compared to endothelial tubes isolated from the same vessels (Socha et al., 2015), EC death induced by H₂O₂ was reduced by more than half. We therefore suggest that greater EC death in endothelial tubes reflects loss of reciprocal protection by the smooth muscle layer. Crosstalk between the intima and media that promotes cell survival is consistent with our finding that endothelial denudation increased SMC death, as did blocking gap junctions in intact arteries. A protective role for myoendothelial coupling through gap junctions is also supported by the lack of effect of carbenoxolone in denuded arteries (Figure 2F). However, SMC death induced by H₂O₂ was elevated in denuded vessels, making it difficult to resolve a significant effect of carbenoxolone. In light of the adverse effects of NO on SMC viability observed here, the mechanism(s) by which myoendothelial signaling mediates reciprocal protection between the intima and media of resistance arteries during acute oxidative stress requires further study.

Advanced age protects SMCs from excessive Ca²⁺ entry and cell death

Ageing increases cardiovascular disease and vascular complications in association with elevated oxidative stress (Alzaid et al., 2014). However, as shown by transitioning from NO to H₂O₂ as an endothelial-derived vasodilator, signaling pathways can adapt during ageing (Muller-Delp et al., 2012;Widlansky & Gutterman, 2011). The present results indicate that advanced age protects both SMCs and ECs during acute exposure to H₂O₂ with a corresponding attenuation of the rise in [Ca²⁺]_i. These findings are corroborated by complementary data from IL-10^−/− mice, providing additional evidence that loss of IL-10 mimics the phenotype of ageing (Walston et al., 2008;Ko et al., 2011). Under the conditions of our experiments, Fura-2 dye added to the bath preferentially enters SMCs for evaluating [Ca²⁺]_i dynamics [(Norton & Segal, 2018) and unpublished observations]. Our previous experiments on endothelial tubes (Socha et al., 2015) verified that Ca²⁺ responses to H₂O₂ are reduced in the endothelium of vessels from old mice; the present data demonstrate that SMCs adapt in a similar manner. During washout of H₂O₂, Ca²⁺ recovered to near baseline levels in SEAs from old and IL-10^−/− mice, while [Ca²⁺]_i remained elevated in vessels from young mice. This observation suggests that the lower [Ca²⁺]_i in the vascular wall of old mice during H₂O₂ exposure may reflect increased Ca²⁺ extrusion through the plasma membrane or greater sequestration by the sarco/endoplasmic reticulum. Further, the [Ca²⁺]_i response to H₂O₂ was nearly abolished in the absence of extracellular Ca²⁺, supporting a role for reduced Ca²⁺ entry as a mechanism of adaptation.

H₂O₂ can activate L-type Ca²⁺ channels (Chaplin et al., 2015) and multiple transient receptor potential (TRP) channels (Andersson et al., 2008;DelloStritto et al., 2016;Sun et al., 2012) expressed by vascular cells. Ageing decreases L-type channel expression and associated membrane currents in mesenteric arteries and aorta (Albarwani et al., 2016;Fukuda et al., 2014), while TRP channel function is impaired in the cerebral vasculature of aged mice (Toth et al., 2013). Thus, a variety of ion channels may contribute to the adaptive response that the present studies have identified for advanced age that is mimicked by the loss of IL-10 and being female in young mice. A limitation to our experimental design is that during the time course of our Ca²⁺-free protocol, intracellular stores may fall and their contribution to the H₂O₂-dependent increases in [Ca²⁺]_i may be underestimated, particularly at later time points. Nevertheless, a small component of the Ca²⁺ response persisted during H₂O₂ exposure. While sarco/endoplasmic reticulum can also be a source of Ca²⁺ that mediates cell death (Ermak & Davies, 2001), our findings collectively illustrate that the major component of H₂O₂-induced increases in [Ca²⁺]_i reflects Ca²⁺ entry from the extracellular fluid.

H₂O₂ induces apoptosis of SMCs

Cell death in the vascular wall can be initiated by a host of signals including cytokines, hormones, and ROS. Among ROS, H₂O₂ plays a key role in apoptosis as it is produced under nearly all conditions of oxidative stress and diffuses freely between cells. Apoptosis is a prominent feature of ischemia/reperfusion injuries (Kalogeries et al., 2016;Granger & Kvietys, 2015) and cardiovascular disease spanning atherosclerosis, hypertension, and stroke (Diez et al., 1998;Peiro et al., 2001;Sugamura & Keaney, 2011). H₂O₂ can induce apoptosis at concentrations ranging from 50 µM to 10 mM (Brunt et al., 2006;Pan & Berk, 2007;Sun et al., 2012;Li et al., 2003). We identified 200 µM as an effective concentration for eliciting cell death within a time frame for studying intact resistance arteries that retained their integrity under control conditions. Furthermore, this concentration of H₂O₂ is consistent with that observed during ischemic events (Hyslop et al., 1995). Propidium iodide crosses disrupted membranes as an index of cell damage and death, however it does not delineate between apoptosis and necrosis. Key events in apoptotic signaling through the mitochondrial (intrinsic) pathway include activation of the Bcl-2 associated proteins BAX or BAK to form pores in the outer mitochondrial membrane (McCarthy et al., 1997;Xiang et al., 1996) resulting in Ca²⁺ overload, loss of mitochondrial membrane potential (Ψ_m) and release of cytochrome C, which combines with apoptotic protease activating factor (APAF)-1 and caspase 9, culminating in activation of caspase 3 to disrupt cellular integrity (Singh et al., 2007;Viola et al., 2007;Lin et al., 2007;Kroemer et al., 2007;Danial & Korsmeyer, 2004;Porter & Janicke, 1999). In addition to these cellular markers (Figure 8), finding that caspase inhibition greatly decreased SMC death indicates that acute exposure to H₂O₂ exposure killed these cells via intrinsic apoptosis.

Endothelial cell death was not induced by H₂O₂ in experiments performed with luminal perfusion, attributable to the ability of shear stress to inhibit apoptosis as shown for human ECs (Dimmeler et al., 1996). Therefore, the increased resilience of ECs to apoptosis induced by ROS may be even greater in vivo. Luminal perfusion may also protect the media, as SMC death was lower when ECs were exposed to flow (Compare Figures 1E and 7C). Although limitations imposed by the adventitia of intact vessels prevented effective use of annexin V to evaluate apoptosis in SMCs, its intermittent staining of ECs when delivered within the vessel lumen supports this interpretation.

Apoptosis following H₂O₂ exposure is documented for isolated vascular cells (Kalyankrishna et al., 2002;Li et al., 2003;Sun et al., 2012). In this study, we evaluated established markers of intrinsic apoptosis for intact vessels. Thus, cytochrome C became diffuse throughout SMCs of vessels exposed to H₂O₂, consistent with release from mitochondria (Figure 8). Oxidative stress increased the active (cleaved) form of caspase 9 and caused it to localize in a manner which may reflect association with the cytoskeleton as a feature of its activation (Amora et al., 2006). Proteolysis of cytoskeletal proteins by caspases is a marker of apoptosis (Kothakota et al., 1997), and such an effect may contribute to the reduction in vasoconstriction to NA that was associated with cell death following H₂O₂ exposure. Similarly, cleaved caspase 3 was increased and became localized in distinct clusters following treatment with H₂O₂, consistent with increased caspase activity as the primary effector of apoptosis.

Conclusion

Perturbations in the balance between cell proliferation and cell death contribute to the pathogenesis of vascular disease (Mallat & Tedgui, 2000;Stefanec, 2000). In contrast to the consensus that ageing increases apoptosis in a variety of cell types (Cooper, 2012), we demonstrate that advanced age (or the absence of IL-10 in young animals) protects SMCs and ECs of intact resistance arteries from H₂O₂-induced apoptosis by reducing Ca²⁺ influx through the plasma membrane. Further, respective cell layers exert reciprocal protection through myoendothelial coupling and preservation of vasomotor function accompanies SMC survival. Collectively, these findings indicate that conditions of chronic oxidative stress increase the resilience of intact resistance arteries to the intrinsic pathway of apoptosis.

TRANSLATIONAL PERSPECTIVE.

Elevated levels of reactive oxygen species (ROS) exert deleterious effects on vascular structure and function and contribute to cardiovascular disease (Diez et al., 1998;Sugamura & Keaney, 2011). Conditions of chronic oxidative stress such as advanced age predispose the vasculature to disease, however, clinical studies have failed to demonstrate significant benefits from antioxidant therapies (Kamel et al., 2006;Cook et al., 2007). Therefore, we challenged the conventional paradigm that chronic oxidative stress exerts only harmful effects to promote cardiovascular disease. We hypothesized that vascular smooth muscle cells (SMCs) and endothelial cells (ECs) adapt during chronic oxidative stress experienced during advanced age to become resilient to cell death induced by acute exposure to hydrogen peroxide. Vascular apoptosis has established roles in hypertensive vascular remodeling, aneurysm, thrombi formation, and disruption of normal blood flow (Clarke et al., 2007). The present data demonstrate that advanced age protects SMCs and ECs from intrinsic apoptosis induced by calcium overload during acute oxidative stress imposed by exposure to hydrogen peroxide. Further, maintaining integrity of the endothelium protects both SMCs and ECs in the vessel wall. Understanding how to evoke such protection irrespective of ageing will provide new insight for maintaining vascular structure and function during acute oxidative stress. These conditions include ischemia/reperfusion injury following organ transplantation, myocardial infarction and ischemic stroke.

KEY POINTS.

• Vascular oxidative stress increases with advancing age. We hypothesized that resistance vessels develop resilience to oxidative stress to protect functional integrity. Isolated pressurized superior epigastric arteries (SEAs; diameter, ~150 μm) of old (24 mo) and young (4 mo) mice were exposed to H₂O₂ (200 μM) for 50 min at 37°C.

• H₂O₂-induced death was greater in smooth muscle cells (SMCs) than endothelial cells (ECs) and lower in SEAs from old vs. young mice. The rise in vessel wall [Ca²⁺]_i induced by H₂O₂ was attenuated with ageing as was the decline in adrenergic vasoconstriction. Genetic deletion of IL-10 mimicked the effects of advanced age on cell survival.

• Inhibiting NO synthase or scavenging peroxynitrite reduced SMC death; endothelial denudation or inhibiting gap junctions increased SMC death. Delocalization of cytochrome C activated caspases 9 and 3 to induce apoptosis.

• Vascular cells develop resilience to H₂O₂ during ageing by preventing Ca²⁺ overload and endothelial integrity promotes SMC survival.

Abbreviations:

EC

endothelial cell

FeTPPS

5, 10, 15, 20-tetrakis (4-sulfonatophenyl) porphyrinato iron III chloride

H₂O₂

hydrogen peroxide

ID

inner diameter

IL-10

interleukin 10

[Ca²⁺]

ntracellular Ca2+

L-NAME

N^G-Nitro-L-arginine methyl ester

NO

nitric oxide

NA

noradrenaline

PI

propidium iodide

ROS

reactive oxygen species

ROI

region of interest

SMC

smooth muscle cell

SEA

superior epigastric artery

Z-VAD-FMK

carbobenzoxy-valyl-alanyl-aspartyl-[O-methyl]- fluoromethylketone