Advanced age protects microvascular endothelium from aberrant Ca²⁺ influx and cell death induced by hydrogen peroxide

Abstract

Abstract

Endothelial cell Ca²⁺ signalling is integral to blood flow control in the resistance vasculature yet little is known of how its regulation may be affected by advancing age. We tested the hypothesis that advanced age protects microvascular endothelium by attenuating aberrant Ca²⁺ signalling during oxidative stress. Intact endothelial tubes (width, ∼60 μm; length, ∼1000 μm) were isolated from superior epigastric arteries of Young (3–4 months) and Old (24–26 months) male C57BL/6 mice and loaded with Fura-2 dye to monitor [Ca²⁺]_i. At rest there was no difference in [Ca²⁺]_i between age groups. Compared to Young, the [Ca²⁺]_i response to maximal stimulation with acetylcholine (3 μm, 2 min) was ∼25% greater in Old, confirming signalling integrity with advanced age. Basal H₂O₂ availability was ∼33% greater in Old while vascular catalase activity was reduced by half. Transient exposure to elevated H₂O₂ (200 μm, 20 min) progressively increased [Ca²⁺]_i to ∼4-fold greater levels in endothelium of Young versus Old. With no difference between age groups at rest, Mn²⁺ quench of Fura-2 fluorescence revealed 2-fold greater Ca²⁺ influx in Young during elevated H₂O₂; this effect was attenuated by ∼75% using ruthenium red (5 μm) as a broad-spectrum inhibitor of transient receptor potential channels. Prolonged exposure to H₂O₂ (200 μm, 60 min) induced ∼7-fold greater cell death in endothelium of Young versus Old. Thus, microvascular endothelium can adapt to advanced age by reducing Ca²⁺ influx during elevated oxidative stress. Protection from cell death during oxidative stress will sustain endothelial integrity during ageing.

Key points

• Calcium signalling in endothelial cells of resistance arteries is integral to blood flow regulation. Oxidative stress and endothelial dysfunction can prevail during advanced age and we questioned how calcium signalling may be affected.

• Intact endothelium was freshly isolated from superior epigastric arteries of Young (∼4 months) and Old (∼24 months) male C57BL/6 mice. Under resting conditions, with no difference in intracellular calcium levels, hydrogen peroxide (H₂O₂) availability was ∼1/3 greater in endothelium of Old mice while vascular catalase activity was reduced by nearly half.

• Compared to Old, imposing oxidative stress (200 μm H₂O₂) for 20 min increased intracellular calcium to 4-fold greater levels in endothelium of Young in conjunction with twice the calcium influx. Prolonged (60 min) exposure to H₂O₂ induced 7-fold greater cell death in endothelium of Young.

• Microvascular adaptation to advanced age may protect endothelial cells during elevated oxidative stress to preserve functional viability of the intima.

Introduction

In the resistance vasculature, the role of the endothelium in governing tissue perfusion is driven by dynamic regulation of intracellular Ca²⁺ concentration [Ca²⁺]_i. For example, elevations in endothelial cell (EC) [Ca²⁺]_i activate small- and intermediate-conductance K_Ca channels (SK_Ca and IK_Ca, respectively) along with production of nitric oxide (NO) and prostacyclin (Crutchley et al. 1983; Kruse et al. 1994; Campbell et al. 1996; Tran & Watanabe, 2006; Ledoux et al. 2008). These complementary signalling events serve to relax the surrounding smooth muscle cells (SMCs), with vasodilatation increasing tissue blood flow. With advanced age, blood flow regulation by the endothelium is characterized by greater levels of reactive oxygen species (ROS) and diminished bioavailability of NO (Taddei et al. 2001; Seals et al. 2011; Muller-Delp et al. 2012). In turn, these changes may impair EC function and compromise tissue perfusion (Feletou & Vanhoutte, 2006; Seals et al. 2011). These adverse effects of advanced age often coincide with a sedentary lifestyle and impaired functional work capacity (Dinenno et al. 1999, 2001; Donato et al. 2006). Of even greater concern, compromising the physiological integrity of the endothelium presages the advance of cardiovascular disease (Yeboah et al. 2007; Lind et al. 2011).

Superoxide generation occurs through mitochondrial respiration, substrate conversion via cytosolic and membrane-associated oxidases, and the uncoupling of nitric oxide synthase (NOS) (Drouin et al. 2007; Widlansky & Gutterman, 2011; Muller-Delp et al. 2012; Bachschmid et al. 2013). While able to react with NO and generate peroxynitrite (Fleming & Busse, 1999; van der Loo et al. 2000; Donato et al. 2007; Trott et al. 2009; Sindler et al. 2013), superoxide anions can be converted rapidly to the stable (and membrane-permeant) ROS hydrogen peroxide (H₂O₂) either spontaneously or via mitochondrial, cytosolic and membrane-associated superoxide dismutases (Widlansky & Gutterman, 2011; Muller-Delp et al. 2012; Bachschmid et al. 2013). While H₂O₂ can contribute to endothelial-dependent dilatation (Miura et al. 2003; Widlansky & Gutterman, 2011; Muller-Delp et al. 2012), high levels of H₂O₂ can elevate EC [Ca²⁺]_i, impair normal function and initiate cell death (Volk et al. 1997; Hecquet et al. 2008; Sun et al. 2012). However, it is unknown how advanced age affects Ca²⁺ responses to H₂O₂ in endothelium of the resistance vasculature, or whether there may be associated changes in EC viability.

In the present study, we tested the hypothesis that advanced age protects microvascular endothelium by attenuating aberrant Ca²⁺ signalling during exposure to elevated H₂O₂. Endothelial tubes freshly isolated from skeletal muscle resistance arteries of Young (3–4 months) and Old (24–26 months) mice were exposed to H₂O₂ at a concentration (200 μm) shown to hyperpolarize ECs to their K⁺ equilibrium potential within ∼20 min through activating K_Ca (Behringer et al. 2013). The present findings reveal that, with no difference in [Ca²⁺]_i under basal conditions, resting levels of H₂O₂ were higher in Old. Exposure to elevated H₂O₂ (200 μm) increased [Ca²⁺]_i to ∼4-fold higher levels and evoked cell death to a ∼7-fold greater extent in endothelium of Young compared to Old, suggesting that microvascular ECs adapt during advanced age to attenuate the rise in [Ca²⁺]_i and minimize cell death when subjected to an abrupt increase in oxidative stress.

Methods

Animal care and use

All procedures were approved by the Institutional Animal Care and Use Committee of the University of Missouri– Columbia. Experiments were performed in accordance with the Guide for the Care and Use of Laboratory Animals of the National Research Council (8th edn, 2011). Young (∼4 months) and Old (∼24 months) male C57BL/6 mice were obtained concurrently from the National Institute on Aging colonies (Charles River Laboratories, Wilmington, MA, USA). Prior to experiments, mice were acclimatized in the same room of the animal care facility at the University of Missouri School of Medicine for at least 1 week on a 12/12 h light–dark cycle at ∼23°C with fresh tap water and standard rodent chow available ad libitum. Young and Old mice were studied in random order under the same conditions within each experimental protocol. On the morning of an experiment, a mouse was anaesthetized with an intraperitoneal injection of pentobarbital sodium (60 mg kg⁻¹) and abdominal fur was removed by shaving. The origin of each superior epigastric artery (SEA) was exposed near the sternum and ligated with 6-0 silk suture. The entire abdominal musculature was then excised and placed in 4°C dissection buffer (pH 7.4, 285–290 mOsm) composed of (in mm): 137 NaCl, 5 KCl, 1 MgCl₂, 10 Hepes, 10 glucose, 0.01 sodium nitroprusside (SNP) and 0.1% bovine serum albumin (BSA) (US Biochemical, Cleveland, OH, USA). Following tissue removal, the anaesthetized mouse was killed via cardiac excision or an overdose of pentobarbital via cardiac injection.

Isolation of endothelial tubes

Intact endothelial tubes were isolated from the SEA as described (Socha et al. 2011, 2012; Socha & Segal, 2013). Briefly, the abdominal musculature was transferred to a dissection chamber at 4°C and pinned onto a transparent silicone rubber block (Sylgard 184; Dow-Corning, Midland, MI, USA). While viewing through a stereomicroscope, each SEA was dissected from the site of ligation to the first major branch point (isolated segment length, 3–3.5 mm) and the lumen was flushed with chilled dissection buffer to remove residual blood. Each SEA was cut into 2–3 smaller segments (length, ∼1–2 mm) and transferred into 12 × 75 mm culture tubes on ice containing 4 ml of dissection buffer. The culture tube was warmed to room temperature (∼24°C) and vessel segments were transferred to a 12 × 75 mm culture tube containing 1 ml of preheated (37°C) dissociation buffer composed of (in mm): 137 NaCl, 5 KCl, 1 MgCl₂, 10 Hepes, 10 glucose, 2 CaCl₂, 0.1% BSA, 0.62 mg ml⁻¹ papain, 1.5 mg ml⁻¹ collagenase and 1.0 mg ml⁻¹ dithioerythritol, pH 7.4. Vessel segments were incubated for 30 min at 37°C then transferred to enzyme-free dissociation buffer at room temperature. A vessel segment was transferred to a recording chamber (RC-27 N; Warner Instruments, Hamden, CT, USA). During visual inspection at 40× magnification, surrounding SMCs and adventitia were dissociated from the endothelium using gentle trituration through the heat-polished tip (internal diameter, ∼120 μm) of a micropipette pulled (P-97; Sutter Instruments, Novato, CA, USA) from borosilicate glass capillary tubes (#1B100-4; World Precision Instruments, Sarasota, FL, USA). The freshly isolated endothelial tube was positioned at the centre of the chamber and aligned along the chamber axis in the direction of fluid flow. Each end was pressed gently against the chamber bottom (a 24 × 50 mm coverslip) using a blunt-tipped pinning pipette (heat-polished borosilicate glass; external diameter, ∼100 μm) secured in three-axis micromanipulators (DT3-100; Siskiyou Design, Grants Pass, OR, USA). The endothelial tube was extended to approximate in situ length by slowly retracting each pinning pipette and superfused (3 ml min⁻¹) with physiological salt solution [PSS, 290–295 mOsm; pH 7.4, composed of (in mm): 137 NaCl, 5 KCl, 1 MgCl₂, 10 Hepes, 10 glucose and 2 CaCl₂].

Calcium photometry

The endothelial tube was equilibrated for ∼15 min during superfusion with PSS at room temperature (∼24°C) then incubated for 30 min without flow in Fura-2 AM dye (# F14185; Invitrogen, Carlsbad, CA, USA). The dye was dissolved in DMSO and diluted to 5 μm in PSS with final DMSO  <  0.5% (Socha et al. 2011). Superfusion with PSS was resumed to wash out excess dye and allow dye within cells to de-esterify while temperature was raised to 32°C during 30 min of equilibration. Calcium photometry was performed with an IonOptix (Milford, MA, USA) system (Socha et al. 2011). Using a 20× objective (Nikon Fluor20; NA 0.75; Nikon Corporation, Tokyo, Japan), Fura-2 was excited alternately (250 Hz) at 340 and 380 nm with a 75 W Xe lamp while fluorescence emission was collected at 510 nm and expressed as the ratio F₃₄₀/F₃₈₀. Auto-fluorescence of the endothelium was recorded prior to dye loading and subtracted from respective recordings. The imaging window for collecting fluorescence emission was 320 μm long and adjusted to accommodate the width of each tube (range, 50–75 μm) while observing it on a video monitor using transmitted light from a halogen lamp (600 nm long pass filter) directed to a charge coupled device camera (LCL-902C; Watec, Orangeburg, NY, USA). To determine minimum and maximum F₃₄₀/F₃₈₀ values in Young and Old mice, tubes were superfused with PSS containing 0 mm Ca²⁺ and 5 mm EGTA for 1 h followed by PSS containing 3 μm ionomycin and 10 mm Ca²⁺, respectively (Socha et al. 2011). Following Fura-2 dye loading and equilibration, an index of basal [Ca²⁺]_i was determined by averaging the F₃₄₀/F₃₈₀ ratio over 30 s under resting conditions. To evaluate [Ca²⁺]_i responses to G-protein coupled receptor (GPCR) activation, 3 μm ACh, which evokes peak responses for [Ca²⁺]_i and hyperpolarization (Behringer et al. 2012, 2013), was added to the superfusion solution for 2 min while the F₃₄₀/F₃₈₀ ratio was recorded. Data collected during the final 30 s of the [Ca²⁺]_i response to ACh were averaged for analysis.

Dichlorofluorescein fluorescence

Endothelial tubes were secured in the recording chamber (see Calcium photometry above). To determine whether basal H₂O₂ availability in endothelium increases with advanced age, cells were loaded with the H₂O₂-sensitive fluorescent probe dichlorofluorescein (DCFH; 6-carboxy-2′,7′-dichlorodihydrofluorescein diacetate; #C-400; Molecular Probes, Eugene, OR, USA). For dye loading, DCFH was dissolved in DMSO and then diluted to 15 μm in PSS (pH 7.4; DMSO < 0.5%). The tube was immersed at room temperature (24°C) for 30 min without flow, then superfusion with PSS (3 ml min⁻¹) was restored to remove excess dye, the signal was allowed to stabilize (1–2 min) and initial baseline fluorescence (F₀) was determined. These experiments were maintained at room temperature because intracellular DCFH levels remained stable at 24°C for at least 30 min. In contrast, DCFH extrusion was pronounced within several min at 32°C (time controls not shown). Utilizing a 40× objective (NA = 0.9, Nikon) and recording from a 140 × 50 μm window, DCFH was excited at 460–500 nm while emission was recorded continuously at 510–560 (505 nm dichroic mirror; Chroma Technology Corp., Bellows Falls, VT, USA) for at least 30 min. Linear regression was performed on the first 10 min, which corresponded to the steepest rise in fluorescence.

Control experiments were performed to confirm the sensitivity of intracellular DCFH to H₂O₂. To inhibit the accumulation of DCFH fluorescence resulting from H₂O₂ produced endogenously, endothelial tubes were pre-incubated in PSS (pH 7.4) containing the superoxide dismutase mimetic TEMPOL (4-hydroxy-2,2,6,6-tetramethylpiperidin-1-oxyl, 1 mm) + PEG-catalase (500 U ml⁻¹) to decompose intracellular H₂O₂ (Behringer et al. 2013). To estimate the [H₂O₂] required to increase DCFH fluorescence above its basal rate of accumulation, H₂O₂ was added to the superfusion solution (pH 7.4) at defined concentrations. Data were acquired using Ion Wizard software (IonOptix) and were analysed after subtraction of background fluorescence.

Catalase activity

To determine whether scavenging of H₂O₂ within the vessel wall is reduced with advanced age, catalase activity was measured using a colorimetric assay. Preliminary experiments determined the amount of vessel required. Thus, for each mouse, first- and second-order superior and inferior epigastric arteries were dissected bilaterally, flushed of blood, then pooled and cut into segments ∼100 μm long. Each pooled sample (one per mouse) was placed in 100 μl lysis buffer consisting of: 50 mm Hepes, 150 mm NaCl, 1 mm MgCl₂, 1% Triton X-100, 2% glycerol and protease inhibitor cocktail (modified from Nystoriak et al. 2011). Samples were vortexed for 1 min, sonicated for 20 min, vortexed again for 1 min, sonicated for another 10 min, then centrifuged at 7000 g for 10 min at 4°C. The lysate supernatant was retained and used for assay. Protein concentrations for each sample were measured using the Micro BCA Assay (#23235; ThermoFisher, Waltham, MA, USA). A standard curve (0–200 μg μl⁻¹ BSA) was prepared and protein concentrations were measured on a NanoDrop 2000c (ThermoFisher). The assay for catalase activity (#CAT100; Sigma-Aldrich, St Louis, MO, USA) was run at room temperature according to the manufacturer's instructions. Standard curves for H₂O₂ concentration were prepared fresh daily and each assay included positive and negative controls in addition to samples from Young and Old mice. Each reaction included 75 μl of lysate supernatant (containing 10–15 μg protein) and lasted 5 min. Assays were read at 520 nm in a quartz cuvette on the NanoDrop 2000c. Specific catalase activity is reported as μmol H₂O₂ consumed min⁻¹ ml⁻¹ (μg protein)⁻¹.

Hydrogen peroxide exposure

To determine whether advanced age impairs Ca²⁺ signalling in microvascular endothelium during acute oxidative stress, endothelial tubes were loaded with Fura-2 dye as described above to evaluate Ca²⁺ responses to H₂O₂. After warming to 32°C, the protocol consisted of an initial baseline recording, addition of ACh (3 μm) to the superfusion solution for 2 min, then washout and recovery to resting baseline (10 min) during superfusion with control PSS. In accordance with the protocol developed for earlier studies (Behringer et al. 2013), H₂O₂ (200 μm) was included in the superfusion solution (pH 7.4) for 20 min followed by 30 min washout and recovery with control PSS. Fura-2 emission was recorded for the final 10 s of each 5 min period during H₂O₂ exposure and throughout recovery with F₃₄₀/F₃₈₀ averaged for each recording. In separate experiments to determine whether Ca²⁺ signals during H₂O₂ exposure were derived from intra- or extracellular sources, the 20 min exposure to H₂O₂ was performed using PSS that was nominally Ca²⁺-free (i.e. prepared without CaCl₂). Superfusion with control PSS (2 mm Ca²⁺) was restored during the 30 min recovery period with Fura-2 emission sampled as above. To determine whether advanced age affected EC viability during oxidative stress, the duration of exposure to H₂O₂ (200 μm) was extended to 60 min.

Manganese quench

Quenching of Fura-2 fluorescence by Mn²⁺ correlates with the rate of divalent cation flux across the plasma membrane, rendering this method an accurate assessment of Ca²⁺ influx (Potocnik & Hill, 2001; Tas et al. 2008). An endothelial tube was loaded with Fura-2 dye as described above and studied at 32°C. To obtain an index of Ca²⁺ influx at rest, the tube was superfused with PSS in which the CaCl₂ was replaced with MnCl₂ for 5 min, during which the peak rate of decrease in the Fura-2 signal was recorded at its isosbestic wavelength (360 nm). To determine how oxidative stress influenced Ca²⁺ influx, the tube was superfused with 200 μm H₂O₂ in PSS containing 2 mm Ca²⁺ for 10 min, then with 200 μm H₂O₂ in nominally Ca²⁺-free PSS for 10 min. As determined in preliminary experiments, this Ca²⁺-free interval prior to Mn²⁺ addition prevented Ca²⁺ influx, thereby avoiding cytosolic Ca²⁺ accumulation and its binding to Fura-2 (which would limit Mn²⁺ binding sites for the quench assay). Ca²⁺-free PSS containing 2 mm MnCl₂ was then added for 5 min during which the Fura-2 signal was recorded. This protocol resulted in measurements of H₂O₂-induced cation permeability coinciding with 20 min exposure to H₂O₂. Linear regression was performed on a 30 s interval during the steepest portion of the Fura-2 response during MnCl₂ exposure (i.e. during maximal Mn²⁺ influx) to determine the maximal rate of fluorescence quenching, thereby providing an index of the peak rate of cation influx across the plasma membrane.

Transient receptor potential (TRP) channel inhibition

To determine whether TRP channels were involved in [Ca²⁺]_i responses, the broad-spectrum TRP channel inhibitor ruthenium red (RuR; Ramsey et al. 2006) was dissolved directly into PSS (5 μm; pH 7.4) and included during superfusion with H₂O₂. Having determined that [Ca²⁺]_i responses to exogenous H₂O₂ in the endothelium of Young were far greater than in Old mice (see Results), endothelial tubes from Young mice were loaded with Fura-2 as described above. After recording baseline F₃₄₀/F₃₈₀ values and control responses to 3 μm ACh, the endothelium was superfused for 20 min with PSS containing 200 μm H₂O₂ + 5 μm RuR followed by control PSS for 30 min of recovery. Fura-2 fluorescence emission was recorded for 10 s at the end of each 5 min period and averaged.

Cell viability assay

To determine whether EC death was responsible for the Fura-2 signal during the 20 min exposure to H₂O₂, a Live/Dead Viability/Cytotoxicity assay for mammalian cells (#L-3224; Life Technologies,; Carlsbad, CA, USA) was performed. Briefly, following isolation, tubes were superfused with 200 μm H₂O₂ for 20 min at 32°C as done for evaluating [Ca²⁺]_i responses. Following a 2 min wash with control PSS while cooling back to 24°C, the Live/Dead assay was performed at room temperature according to the manufacturer's recommendations. Thus, individual tubes were incubated for 30 min in PSS containing 2 μm calcein-AM (hydrolysed and retained in viable cells) and 4 μm ethidium bromide (EtBr; stains nuclei of dead and dying cells). The tube was washed again for 2 min in control PSS. As there was negligible EtBr staining following 20 min of H₂O₂ for either age group (see Results), additional experiments were performed in which the duration of H₂O₂ exposure was extended to 60 min prior to performing the Live/Dead assay. Images of each tube were acquired using a 40× oil immersion objective (NA = 1.25) on a Leica SP5 laser scanning confocal microscope equipped with LAS Software (Leica Microsystems, Buffalo Grove, IL, USA). Both dye molecules were excited with the 488 nm line of an argon laser with emission acquired at 500–535 nm (calcein) and 595–700 nm (EtBr). Transmitted light images were acquired concurrently. To quantify cell damage in each tube, the number of ECs stained with EtBr was expressed as a percentage of the total number of fluorescently labelled ECs [i.e. (EtBr-positive cells)/(EtBr-positive cells + calcein-positive cells)].

Data analysis and statistics

One EC tube was studied from each SEA. Unpaired Student's t-tests were used to evaluate differences between age groups in F₃₄₀/F₃₈₀ values at rest, in response to ACh and during Mn⁺² quench, during DCFH fluorescence and for catalase activity. Changes in F₃₄₀/F₃₈₀ over time during H₂O₂ exposure were analysed using two-way ANOVA with Bonferroni post-tests. Statistical analyses were performed using GraphPad Prism (GraphPad Software, La Jolla, CA, USA). Summary data are presented as means ± SEM. Differences were accepted as statistically significant at P < 0.05.

Results

Calcium photometry

Time controls confirmed that baseline F₃₄₀/F₃₈₀ did not change significantly for at least 2 h in either Young or Old mice (n = 4 each). Thus, consistent with earlier findings (Socha et al. 2011), resting [Ca²⁺]_i remained stable at 32°C. To substantiate comparisons of F₃₄₀/F₃₈₀ between Young and Old, minimum and maximum F₃₄₀/F₃₈₀ values for the Ca²⁺ signal were determined as described previously for endothelial tubes (Socha et al. 2011). Thus, minimum F₃₄₀/F₃₈₀ values were 0.61 ± 0.02 for Young (n = 3) and 0.64 ± 0.02 for Old (n = 4) with respective maximum F₃₄₀/F₃₈₀ values of 5.13 ± 0.76 and 5.92 ± 1.35; neither value was significantly different between age groups (P = 0.44 and 0.61, respectively). With F₃₄₀/F₃₈₀ as an index of [Ca²⁺]_i, resting values were not different between endothelium of Young versus Old mice (Fig.1A). In response to stimulation with ACh (3 μm), the mean steady-state [Ca²⁺]_i response was 25% higher in Old versus Young (P < 0.05) (Fig.1C), confirming the integrity of signalling through muscarinic GPCRs in advanced age.

Figure 1.

Enhanced [Ca²⁺]_i response to ACh in microvascular endothelium with advanced age

A, resting F₃₄₀/F₃₈₀ ratios were not different between endothelium of Young versus Old mice indicating similar [Ca²⁺]_i at rest. B, representative F₃₄₀/F₃₈₀ recordings from endothelium of Young and Old in response to 3 μm ACh (added at time 0). C, maximal change in [Ca²⁺]_i (ΔF₃₄₀/F₃₈₀) of endothelium from Young and Old mice during sustained response to 3 μm ACh. Summary data are means ± SEM; n = 33 for Young, n = 31 for Old. *P < 0.05, Old versus Young.

Hydrogen peroxide availability under basal conditions

Baseline raw fluorescence (F₀) units for DCFH at the onset of recordings were 474 ± 32 for Young and 402 ± 34 for Old (means ± SEM, n = 10 per age group) and not significantly different between age groups (P = 0.15). To provide an internal control for variability in dye loading and other factors that may influence F₀, these data are presented relative to F₀ (i.e. F/F₀). The rate of DCFH fluorescence accumulation was 33% greater in endothelium of Old versus Young (P < 0.05; Fig.2A–C). In the presence of TEMPOL (1 mm) + PEG-catalase (500 U ml⁻¹) DCFH fluorescence did not change over time (Fig.2A–C), confirming the applicability of this probe for detecting H₂O₂ in our experiments. In lysates of epigastric arteries, endogenous catalase activity was lower by nearly half in Old compared to Young (P < 0.05, Fig.2D). Thus, microvascular endothelium of Old mice exhibits greater H₂O₂ availability under basal conditions in conjunction with diminished capacity of the vessel wall to remove H₂O₂ when compared to Young. To provide an index of H₂O₂ availability under basal conditions, H₂O₂ was added to the superfusion solution. In endothelial tubes of Young (n = 3), the slope of fluorescence accumulation was unaffected by 1 μm H₂O₂ but began to increase following addition of 5 μm H₂O₂. For reference, 200 μm H₂O₂ promptly increased DCFH fluorescence and saturated the signal within 2–3 min (data not shown).

Figure 2.

Enhanced H₂O₂ availability with diminished H₂O₂ decomposition during advanced age

A, representative recordings of DCFH in endothelium of Young mice illustrating progressive increase in fluorescence over 10 min during superfusion with PSS containing 2 mm Ca²⁺ that was prevented by treatment with TEMPOL (1 mm, superoxide dismutase mimetic) + PEG-catalase (500 U ml⁻¹, H₂O₂ scavenger). B, as in A for Old mice; note greater rate of fluorescence accumulation versus Young. C, summary data for rates of fluorescence accumulation in endothelium of Young and Old mice determined from recordings as represented by those in A and B. Rates of fluorescence accumulation were depressed by TEMPOL + PEG-catalase to values not different between Young and Old (n = 10 per age group with PSS; n = 4 per group for TEMPOL + PEG-catalase). *P < 0.05, Old versus Young, ^#P < 0.05, TEMPOL + PEG-catalase versus PSS for Young; ^ΨP < 0.05, TEMPOL + PEG-catalase versus PSS for Old. D, catalase activity measured via H₂O₂ decomposition by lysates of epigastric arteries from Young and Old mice (n = 5 per age group). *P < 0.05, Old versus Young. Summary data in C and D are means ± SEM.

Effect of H₂O₂ on endothelial Ca²⁺ and cellular integrity

Exposing the endothelium of Young mice to 200 μm H₂O₂ for 20 min progressively increased [Ca²⁺]_i. After a delay of several min, the increase was significant by 20 min (Fig.3C) and was nearly twice that observed in response to maximal stimulation with 3 μm ACh (Fig.1C). During the ensuing washout of H₂O₂, this aberrant elevation in [Ca²⁺]_i persisted for the initial 5 min before recovering to control levels within 20–30 min (Fig.3A). In endothelium of Old mice, exposure to 200 μm H₂O₂ also led to a rise in [Ca²⁺]_i with a time course similar to that for Young (Fig.3B), although the [Ca²⁺]_i response was far less (∼25%; P < 0.05) than observed for Young. In separate experiments, omitting Ca²⁺ from the superfusion solution abolished the response to H₂O₂ in both Young (Fig.3A) and Old (Fig.3B) when compared to responses in the presence of 2 mm [Ca²⁺]_o. Within 5 min of reintroducing 2 mm Ca²⁺ during H₂O₂ washout, [Ca²⁺]_i increased in endothelium of Young (Fig.3A) but not in Old (Fig.3B), further indicating a persistent effect of H₂O₂ on Ca²⁺ entry.

Figure 3.

Reduced [Ca²⁺]_i response of microvascular endothelium to H₂O₂ with advanced age

A, exposing endothelium from Young mice to 200 μm H₂O₂ for 20 min progressively increased [Ca²⁺]_i (ΔF₃₄₀/F₃₈₀). [Ca²⁺]_i recovered during 30 min washout of H₂O₂ with PSS containing 2 mm Ca²⁺. Removing extracellular Ca²⁺ (0 [Ca²⁺]_o) during H₂O₂ exposure prevented the rise in [Ca²⁺]_i in endothelial tubes of Young mice yet [Ca²⁺]_i increased within 5 min of H₂O₂ washout with PSS containing 2 mm Ca²⁺. B, as in A for Old mice; note diminished [Ca²⁺]_i response compared to Young in the presence of 2 mm [Ca²⁺]_o and negligible [Ca²⁺]_i response in 0 [Ca²⁺]_o. Summary data are means ± SEM. For both age groups n = 9 for 2 mm [Ca²⁺]_o during H₂O₂, and n = 7 for 0 [Ca²⁺]_o during H₂O₂. *P < 0.05, versus Control. ^#P < 0.05, Old versus Young for given time and condition, ^ΨP < 0.05 for 0 [Ca²⁺]_o versus 2 mm [Ca²⁺]_o exposed to 200 μm H₂O₂.

The above data implicate a key role for extracellular Ca²⁺ in the [Ca²⁺]_i response during elevated H₂O₂. Thus, Ca²⁺ influx across the plasma membrane was evaluated through measuring Fura-2 quench by Mn⁺² (Fig.4). Consistent with the similarity in resting [Ca²⁺]_i, there was no difference in the quench rate of Fura-2 fluorescence at rest between endothelium of Young versus Old mice (Fig.4A and B). During exposure to 200 μm H₂O₂, the quench rate increased for endothelium of Young and Old; however, the rate was ∼2-fold greater (P < 0.05) in Young (Fig.4C and D). Given that TRP channels encompass redox-sensitive cation channels in the plasma membrane of ECs (Poteser et al. 2006; Xu et al. 2008; Sun et al. 2012) and are integral to Ca²⁺ influx (Ramsey et al. 2006), we tested whether H₂O₂-induced Ca²⁺ influx was mediated by TRP channels. In endothelium of Young, inclusion of 5 μm RuR [a broad-spectrum blocker of TRP channels (Ramsey et al. 2006)] during exposure to 200 μm H₂O₂ abrogated the [Ca²⁺]_i response by ∼70% (Fig.5).

Figure 4.

Enhanced Mn²⁺ quenching of Fura-2 fluorescence during H₂O₂ is reduced with advanced age

A, representative Fura-2 recording at isosbestic wavelength (360 nm) during superfusion with 2 mm MnCl₂ (replacing CaCl₂ in PSS) of endothelium from Young and Old mice at rest. Note similar rates of Fura-2 quench. B, summary data (means ± SEM) indicate no difference in Mn²⁺ influx between endothelium of Young (n = 6) and Old (n = 8) mice. C, as in A during exposure to 200 μm H₂O₂. Note increased rate of Fura-2 quench in both age groups compared to A with greater rate in Young versus Old. D, summary data (means ± SEM) indicate greater rate of Mn²⁺ influx in endothelium of Young (n = 6) versus Old (n = 6) during exposure to 200 μm H₂O₂. *P < 0.05, Old versus Young.

Figure 5.

TRP channels contribute to Ca²⁺ influx induced by H₂O₂

In endothelium of Young mice, inclusion of 5 μm RuR to inhibit TRP channels reduced influx of Ca²⁺ during exposure to 200 μm H₂O₂ by ∼75%. This effect was reversed within 5 min of washout of H₂O₂ with PSS. Summary data are means ± SEM; n = 9 for 2 mm [Ca²⁺]_o during H₂O₂ and n = 5 for 5 μm RuR during H₂O₂. *P < 0.05, Old versus Young.

Elevated levels of H₂O₂ can damage ECs (Volk et al. 1997; Sun et al. 2012). To determine whether H₂O₂ exposure was associated with differences in cellular integrity between age groups, Live/Dead cell assays were performed. Following the 20 min exposure to 200 μm H₂O₂, cells remained viable in the endothelium of both age groups, as confirmed by their uptake and retention of calcein-AM and absence of staining with EtBr (n = 3 per age group; data not shown). Increasing the duration of H₂O₂ exposure to 60 min resulted in dead and dying cells within endothelial tubes of both age groups (Fig.6B and C). Remarkably, there was a pronounced protective effect of advanced age. Thus, with 566 ± 139 cells counted per endothelial tube in Young (n = 5) and 616 ± 49 cells counted per endothelial tube in Old (n = 5), the percentage of ECs labelled with EtBr was ∼7-fold greater (P < 0.05) in Young compared to Old (Fig.6E). Consistent with the stability of Ca²⁺ and electrical responses in our previous studies (Socha et al. 2011; Behringer et al. 2012, 2013), time controls performed here confirmed that, in the absence of H₂O₂, there were no dead or dying cells in endothelial tubes of Young following 60 min under identical conditions (n = 2).

Figure 6.

Advanced age abrogates cell death induced by prolonged exposure to H₂O₂

Following 60 min exposure to 200 μm H₂O₂, a Live/Dead cell assay was performed on endothelial tubes from Young (left) and Old (right) mice. A, living endothelial cells sequestered and retained calcein-AM (green), confirming their viability. B, dead and dying cells (which have nuclei stained red with EtBr) appear more frequently for endothelium of Young versus Old. C, overlay of images in A and B. D, brightfield images of endothelial tubes in A–C. E, the percentage of EtBr-positive cells is ∼7-fold greater in endothelium of Young versus Old. Summary data are means ± SEM; n = 5 per age group. *P < 0.05, Old versus Young. Scale bars = 50 μm.

Discussion

Calcium signalling is integral to endothelial function and vasomotor control. However, there is a paucity of information regarding the effects of advanced age on endothelial Ca²⁺ signalling during oxidative stress, particularly in the microcirculation. To address this gap in understanding, we evaluated [Ca²⁺]_i at rest, during muscarinic GPCR activation and in response to acute oxidative stress with H₂O₂ (200 μm) in endothelium freshly isolated from skeletal muscle resistance arteries of Young and Old mice. Resting [Ca²⁺]_i was not different between age groups, yet the peak steady-state [Ca²⁺]_i response to ACh (3 μm) was greater in endothelium of Old compared to Young. Furthermore, when compared to Young, the constitutive release of H₂O₂ under basal conditions was greater for endothelium of Old and was accompanied by lower vascular catalase activity. During oxidative stress imposed by exposure to H₂O₂ for 20 min, the endothelium of Young mice responded with an aberrant increase in [Ca²⁺]_i whereas the endothelium of Old mice did not. As indicated by its dependence on extracellular Ca²⁺, this difference between age groups may reflect adaptation of microvascular endothelium during advanced age that attenuates influx of Ca²⁺ across the plasma membrane. Moreover, prolonged (60 min) exposure to H₂O₂ resulted in far greater cell death in endothelium of Young than in Old. These findings demonstrate a protective effect of advanced age on Ca²⁺ influx and cell viability in response to acute oxidative stress in native endothelium of resistance microvessels.

Endothelial dysfunction (e.g. impaired endothelial-dependent dilatation, EDD) accompanies advancing age in humans (Taddei et al. 2001; Seals et al. 2011) and is a risk factor for cardiovascular disease (Yeboah et al. 2007; Lind et al. 2011). In the peripheral circulation, the efficacy of endothelium-dependent vasodilators (e.g. ACh and methacholine; which act through muscarinic GPCRs to elevate [Ca²⁺]_i) is diminished in elderly humans (e.g. > 60 years) compared to their younger (e.g. < 30 years) counterparts (Taddei et al. 2001). At the same time, endothelium-independent vasodilatation (e.g. to the NO donor SNP) was maintained (Taddei et al. 2001). Remarkably, little is known of how ageing may affect endothelial [Ca²⁺]_i responses to GPCR agonists or to acute oxidative stress. Therefore, we tested the hypothesis that advanced age impairs endothelial Ca²⁺ signalling. Under resting conditions, [Ca²⁺]_i was not different between endothelial tubes from Young versus Old mice (Fig.1A). Through binding to muscarinic GPCRs, ACh initiates intracellular Ca²⁺ release from the endoplasmic reticulum with an ensuing prolonged phase of Ca²⁺ entry across the plasma membrane from the extracellular fluid (Tran & Watanabe, 2006; Socha et al. 2011). Our use of 3 μm ACh as a maximal criterion stimulus is based upon earlier studies of endothelial tubes, where ACh elevated [Ca²⁺]_i and evoked hyperpolarization through K_Ca activation in a concentration-dependent manner. Both [Ca²⁺]_i and hyperpolarization plateaued at ≥ 1 μm ACh (Behringer et al. 2012), with no difference in the magnitude of hyperpolarization between Young and Old (Behringer et al. 2013). During the prolonged phase studied here, elevations in [Ca²⁺]_i were ∼33% greater in endothelium of Old versus Young (Fig.1C). This finding confirms the integrity of this signalling pathway while suggesting augmented Ca²⁺ entry with advanced age. Greater elevation of [Ca²⁺]_i during sustained muscarinic GPCR activation may help to compensate for age-related reductions in NO bioavailability (Muller-Delp, 2006; Seals et al. 2011). Thus, elevated [Ca²⁺]_i would promote the activation of endothelial NOS. However, elevated [Ca²⁺]_i can enter mitochondria, thereby increasing mitochondrial generation of ROS (Camello-Almaraz et al. 2006) and EC oxidative stress.

Augmented superoxide production and greater levels of oxidative stress have been demonstrated in ECs with ageing (Dhalla et al. 2000; Muller-Delp et al. 2012; Bachschmid et al. 2013). Elevation of ROS occurs as a consequence of greater synthesis, diminished removal or a combination thereof. Sources of superoxide (i.e. the precursor to H₂O₂) in ECs include NADPH oxidase, xanthine oxidase, mitochondria (i.e. respiratory enzymes) and uncoupled NOS (Lambeth, 2004; Forstermann, 2008; Touyz et al. 2011). In addition to constitutive ROS production, ECs are exposed to ROS released by surrounding cells (e.g. SMCs) and by cells carried in the bloodstream (Li & Shah, 2004; Feletou & Vanhoutte, 2006). Once generated, superoxide anions form H₂O₂ via the superoxide dismutase family of enzymes. Catalase then decomposes H₂O₂ by converting it to H₂O and O₂ to protect cells from oxidative damage. Levels of oxidative stress, as well as subsequent cellular responses, can vary between vascular beds (Grinnell et al. 2012). To determine whether oxidative stress was manifest under basal conditions within microvascular ECs during advanced age, we evaluated rates of H₂O₂ production using the H₂O₂-sensitive dye DCFH (Carter et al. 1994) after confirming its sensitivity to H₂O₂ under the conditions of our experiments. Our data illustrate that the availability of H₂O₂ was significantly greater in the endothelium of Old mice under resting conditions (Fig.2C). This finding helps to explain the enhanced activation of SK_Ca and IK_Ca observed for endothelial tubes of Old mice (Behringer et al. 2013). With no difference in [Ca²⁺]_i at rest (Fig.1A), direct activation and/or enhanced sensitivity to Ca²⁺ of K_Ca is implied, as channel expression was not different between age groups (Behringer et al. 2013). Complementary experiments revealed that catalase activity was reduced by nearly half in epigastric arteries of Old mice (Fig.2D), supporting higher levels of constitutive oxidative stress in microvascular ECs during advanced age.

Acute elevation of oxidative stress was imposed by superfusion of endothelial tubes with 200 μm H₂O₂ for 20 min. This concentration of H₂O₂ and duration of exposure were based upon finding that they effectively activated SK_Ca and IK_Ca in the same preparations studied here (Behringer et al. 2013). Moreover, the time course observed for the progression of hyperpolarization is consistent with that found here for the rise in [Ca²⁺]_i; i.e. an initial delay upon exposure followed by a progressive increase during the 20 min exposure (Fig.3). Observing that the effect of H₂O₂ on [Ca²⁺]_i persisted during the early period of washout suggests that, despite being membrane-permeant, the actions of H₂O₂ are not instantaneous and that they exert a residual effect following H₂O₂ removal. Compared to the gradual accumulation of DCFH fluorescence under basal conditions, the fluorescence signal saturated within 2–3 min upon exposure to 200 μm H₂O₂. Thus 200 μm H₂O₂ is far greater than levels present under basal conditions. As exposure to 5 μm H₂O₂ was found to increase DCFH fluorescence above basal accumulation, we suggest that this far lower concentration more closely approximates constitutive levels of H₂O₂ availability.

Regulation of EC [Ca²⁺]_i occurs though multiple signalling pathways which include stimulation of GPCRs with physiological agonists (Tran & Watanabe, 2006). Thus, ACh binding to muscarinic (M3) receptors generates inositol 1,4,5 trisphosphate, thereby stimulating releasing of Ca²⁺ from the endoplasmic reticulum followed by sustained influx through the plasma membrane (Tran & Watanabe, 2006). Free Ca²⁺ is then removed from the cytoplasm by ATPases in respective membranes. Indeed, the maintenance of [Ca²⁺]_i homeostasis is essential for EC function and survival, with aberrant levels of [Ca²⁺]_i  leading to dysfunction and cell death (Orrenius et al. 2003; Berridge, 2005). Exposure to 200 μm H₂O₂ resulted in an ∼4-fold greater increase in [Ca²⁺]_i for endothelium of Young (Fig.3A) compared to Old (Fig.3B). To evaluate the role of Ca²⁺ entry, extracellular Ca²⁺ was removed during H₂O₂ treatment, which effectively prevented the increase in [Ca²⁺]_i. That H₂O₂ acted via augmenting Ca²⁺ entry was tested further by using Mn²⁺ influx to quench Fura-2 (Potocnik & Hill, 2001; Tas et al. 2008). Consistent with resting [Ca²⁺]_i, we found no difference in the rate of Fura-2 fluorescence quenching between endothelium of Young versus Old at rest (Fig.4A and B). However, exposure to 200 μm H₂O₂ evoked ∼2-fold greater peak rates of Fura-2 fluorescence quenching for endothelium of Young versus Old (Fig.4C and D). These findings indicate that elevated [Ca²⁺]_i in Young during H₂O₂ exposure can be explained by enhanced Ca²⁺ influx. In light of greater Ca²⁺ influx during ACh exposure in endothelium of Old versus Young mice, the corresponding reduction in Ca²⁺ influx during 200 μm H₂O₂ exposure in Old mice implies an adaptation that occurs in association with significantly higher basal levels of oxidative stress during advanced age.

TRP channels are Ca²⁺-permeant, expressed in ECs and sensitive to oxidative stress (Poteser et al. 2006; Yoshida et al. 2006; Xu et al. 2008; Beech, 2013). In accordance with the far greater Ca²⁺ responses for endothelium of Young versus Old mice, we investigated whether TRP channels were involved in the H₂O₂-induced Ca²⁺ influx in endothelium of Young mice. Remarkably, inclusion of the broad-spectrum TRP channel antagonist RuR (Ramsey et al. 2006) attenuated the H₂O₂-induced [Ca²⁺]_i rise (Fig.5) to levels similar to those observed in the absence of [Ca²⁺]_o (Fig.3). Multiple isoforms of TRP channels are redox sensitive through activation by oxidative or reductive agents. Thus, TRPC5 can be activated directly via sulfhydryl modification by NO and other oxidative agents (Yoshida et al. 2006) or by reducing agents such as thioredoxin (Xu et al. 2008). Furthermore, TRPV1 and TRPA1 can be activated by oxidizing agents through cysteine modification (Yoshida et al. 2006; Macpherson et al. 2007). As recently shown for endothelium of rat cerebral arteries, generation of H₂O₂ leads to lipid peroxidation products which activate ‘TRPA1 sparklets’ of Ca²⁺ influx (Sullivan et al. 2015). In addition to these homomeric channels, TRP channel protein isoforms can form redox-sensitive heteromeric channels. For example, association of TRPC3 and TRPC4 subunits have been found to form such channels in ECs (Poteser et al. 2006). The TRPM2 channel, which is robustly expressed in ECs, is activated by H₂O₂ and the resulting Ca²⁺ influx can lead to apoptosis (Hecquet et al. 2008; Sun et al. 2012). Owing to the possibility that the H₂O₂-induced Ca²⁺ influx could lead to cell death, we performed Live/Dead cell assays and found no evidence thereof following 20 min exposure to 200 μm H₂O₂. This outcome is in agreement with our earlier studies demonstrating that endothelial tubes remained viable (e.g. membrane potential was maintained) for this period under the same experimental conditions (Behringer et al. 2013). However, when the duration of exposure to 200 μm H₂O₂ was increased to 60 min, cell death was apparent in both Young and Old (Fig.6). Remarkably, the prevalence of dead and dying cells (EtBr positive) observed in Young was ∼7-fold greater (∼36%) than that in Old (∼5%). In turn, we suggest that the profound reduction in Ca²⁺ influx observed for endothelium of Old versus Young reflects a protective adaptation during advanced age that is associated with constitutively greater levels of H₂O₂ under basal conditions (Fig. 2).

The production of ROS within ECs can disrupt signalling pathways and lead to cytotoxicity (Dhalla et al. 2000). For example, superoxide can react with NO to form peroxynitrite and thereby reduce the bioavailability of NO while uncoupling NOS (Landmesser et al. 2003), contributing further to superoxide production. In diabetes, endothelial dysfunction induced by oxidative stress results from the activation of protein kinase C and an increase in glycation end-product formation (Brownlee, 2001). Additionally, NADPH oxidase-induced oxidative stress is a key step in the progression of hypertension (Laursen et al. 1997; Schnackenberg et al. 1998) while mounting evidence implicates ROS in the pathogenesis of atherosclerosis (Madamanchi & Runge, 2007). Furthermore, constitutively elevated H₂O₂ during advanced age impairs electrical signal transmission along the endothelium via activation of K_Ca channels, with ‘leaky’ membranes dissipating charge transfer from cell to cell (Behringer et al. 2013).

Scavenging ROS has been found to ameliorate endothelial dysfunction and promote blood flow to skeletal muscle in both humans (Kirby et al. 2009) and mice (Fleenor et al. 2012) of advanced age. Nevertheless, endothelial-derived ROS have increasingly been found to play a key role in cellular signalling pathways under physiological conditions (Ungvari et al. 2006; Thomas et al. 2008; Suvorava & Kojda, 2009; Wolin, 2009; Muller-Delp et al. 2012). Specifically, H₂O₂ has been identified as an ideal second messenger amongst ROS due to its relatively long half-life compared to superoxide or peroxynitrite. Furthermore, unlike the latter anions, which are membrane-impermeant and react in their vicinity of production, its lack of charge enables H₂O₂ to readily permeate cell membranes and thereby serve as an autacoid. Indeed, H₂O₂ has been shown to function as an endothelial-derived vasodilator. Thus, in mesenteric arteries of both humans and mice, endothelial-derived H₂O₂ can hyperpolarize and thereby relax SMCs (Matoba et al. 2000, 2002) to produce vasodilatation. In human coronary arterioles, H₂O₂ generated from mitochondria can regulate flow-mediated (i.e. endothelium-dependent) dilatation through activating large-conductance K_Ca channels of SMCs (Liu et al. 2003). Additionally, recent studies of first-order arterioles from rat soleus muscle indicate that with advanced age there is a shift in EDD from reliance on NO to a greater role for H₂O₂ (Sindler et al. 2013). Such a shift during ageing would imply an adaptation in the endothelium to reduce the cellular damage otherwise caused by H₂O₂, which is consistent with the present observations.

Summary and perspective

The present study has investigated Ca²⁺ signalling in endothelial tubes freshly isolated from resistance arteries of mouse skeletal muscle with respect to the effects of advanced age and acute oxidative stress induced by acute exposure to elevated H₂O₂. Utilization of this model enables studies of native intact microvascular endothelial function in the absence of confounding interactions with circulating blood, surrounding smooth muscle and adventitia. Our findings indicate that microvascular endothelium of Old mice has higher levels of H₂O₂ availability under basal conditions. With greater activation of K_Ca (Behringer et al. 2013) and no difference in [Ca²⁺]_i, at rest in Old versus Young, the present data suggest that constitutive production of H₂O₂ activates K_Ca of native microvascular endothelium, consistent with the effects of H₂O₂ in human endothelial cells (Bychkov et al. 1999; Miura et al. 2003). Exposure to 200 μm H₂O₂ for 20 min evoked far greater [Ca²⁺]_i increases in the endothelium of Young versus Old mice with a time course similar to that observed for hyperpolarization (Behringer et al. 2013). This rise in [Ca²⁺]_i can be explained by greater Ca²⁺ influx across the plasma membrane and inhibiting this effect by blocking TRP channels further implicates the activation of membrane ion channels by H₂O₂. Remarkably, cell injury and death following prolonged (60 min) exposure to H₂O₂ were far greater in endothelium of Young mice compared to Old. Thus, constitutively elevated basal levels of oxidative stress during advanced age may lead to an adaptive response within the endothelium of resistance microvessels that limits aberrant Ca²⁺ influx and mitigates the cytotoxic effects of oxidative stress (Dhalla et al. 2000) in association with greater utilization of H₂O₂ as a mediator of EDD (Widlansky & Gutterman, 2011; Muller-Delp et al. 2012). Indeed, such protective adaptations to advanced age could explain why antioxidant therapy in humans has little beneficial effect and may instead act to increase patient morbidity and mortality (Yoshihara et al. 2010). Our findings collectively indicate a delicate balance between maintaining physiological levels of ROS during normal vascular function (e.g. ‘healthy ageing’) and pathological oxidative stress with associated endothelial dysfunction.

Glossary

DCFH

6-carboxy-2′,7′-dichlorodihydrofluorescein diacetate

EDD

endothelial-dependent dilatation

EC

endothelial cell

EtBr

ethidium bromide

GPCR

G-protein coupled receptor

IK_Ca

intermediate conductance calcium-activated potassium channel

K_Ca

calcium-activated potassium channel

NO

nitric oxide

NOS

nitric oxide synthase

PSS

physiological salt solution

ROS

reactive oxygen species

RuR

ruthenium red

SK_Ca

small conductance calcium-activated potassium channel

SMC

smooth muscle cell

SNP

sodium nitroprusside

SEA

superior epigastric artery

TEMPOL

4-hydroxy-2,2,6,6-tetramethylpiperidin-1-oxyl

TRP

transient receptor potential

Additional information

Competing interests

The authors declare no competing interests.

Author contributions

M.J.S. designed and performed the experiments using Fura-2 dye, analysed and interpreted the data, drafted the manuscript and prepared the figures. E.M.B. evaluated catalase activity, contributed to experimental design, data interpretation and edited the manuscript. R.L.S. evaluated H₂O₂ levels utilizing DCFH and edited the manuscript. E.J.B., T.L.D. and S.S.S contributed to experimental design, analysis, interpretation and presentation of data, and edited the manuscript. All authors reviewed and approved the final version of the manuscript for publication.

Funding

This research was supported by NIH grants R37-HL041026 and R01-HL086483 to S.S.S., F32-HL107050 to M.J.S., F32-HL118836 to E.M.B., K99-AG047198 to E.J.B. and K01-AG041208 to T.L.D. The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Author's present address

M. J. Socha: Biology Department, Loyola Science Center, 204 Monroe Avenue, The University of Scranton, Scranton, PA 18510, USA.