Effect of Hydrogen Peroxide and Superoxide Anions on Cytosolic Ca2+: Comparison of Endothelial Cells from Large-Sized and Small-Sized Arteries

We compared the Ca2+ responses to reactive oxygen species (ROS) between mouse endothelial cells derived from large-sized arteries, aortas (aortic ECs), and small-sized arteries, mesenteric arteries (MAECs). Application of hydrogen peroxide (H2O2) caused an increase in cytosolic Ca2+ levels ([Ca2+]i) in both cell types. The [Ca2+]i rises diminished in the presence of U73122, a phospholipase C inhibitor, or Xestospongin C (XeC), an inhibitor for inositol-1,4,5-trisphosphate (IP3) receptors. Removal of Ca2+ from the bath also decreased the [Ca2+]i rises in response to H2O2. In addition, treatment of endothelial cells with H2O2 reduced the [Ca2+]i responses to subsequent challenge of ATP. The decreased [Ca2+]i responses to ATP were resulted from a pre-depletion of intracellular Ca2+ stores by H2O2. Interestingly, we also found that Ca2+ store depletion was more sensitive to H2O2 treatment in endothelial cells of mesenteric arteries than those of aortas. Hypoxanthine-xanthine oxidase (HX-XO) was also found to induce [Ca2+]i rises in both types of endothelial cells, the effect of which was mediated by superoxide anions and H2O2 but not by hydroxyl radical. H2O2 contribution in HX-XO-induced [Ca2+]i rises were more significant in endothelial cells from mesenteric arteries than those from aortas. In summary, H2O2 could induce store Ca2+ release via phospholipase C-IP3 pathway in endothelial cells. Resultant emptying of intracellular Ca2+ stores contributed to the reduced [Ca2+]i responses to subsequent ATP challenge. The [Ca2+]i responses were more sensitive to H2O2 in endothelial cells of small-sized arteries than those of large-sized arteries.


Introduction
Vascular endothelial cells in vivo are constantly exposed to ROS that are released from neutrophils, macrophages, and vascular smooth muscle cells [1,2]. Moreover, endothelial cells themselves are generators of ROS [1,2]. The main ROS that are produced include superoxide anions, H 2 O 2 , hydroxyl radicals and peroxynitrite. Functionally, ROS play a key role in physiological and pathological processes in endothelial cells. For example, H 2 O 2 at physiological concentration serves as an endothelium-derived hyperpolarizing factor (EDHF), mediating vascular relaxation [3]. However, excessive production of ROS causes extensive damage to the structure and function of endothelial cells, leading to endothelial dysfunction [1]. Evidence indicates that ROSinduced endothelial function and dysfunction are often preceded by an alteration in endothelial [Ca 2+ ] i [4], which serves as an important second messenger to induce diverse responses.
Although there have been a great number of studies investigating the ROS effect on [Ca 2+ ] i in endothelial cells, most of these reports only investigated the endothelial cells derived from large-sized arteries [5][6][7][8][9][10]12,15]. The role of ROS on [Ca 2+ ] i in endothelial cells of small-sized arteries has received little attention [but see 8]. It is unclear whether there is any difference in ROSinduced [Ca 2+ ] i responses in endothelial cells from different-sized arteries. Large-sized arteries and small-sized arteries differ in their function. Small-sized arteries such as mesenteric arteries are resistance arteries that play a key role in blood pressure control. Vasoactive factors in small-sized arteries are often different from that in large-sized arteries. For example, while nitric oxide is the major vasodilator in large arteries, EDHFs often play a more important role as vasodilators in small-sized arteries [16].
In It is well documented that IP 3 -sensitive Ca 2+ stores are the major intracellular Ca 2+ stores, and that the Ca 2+ release from the stores hinges on the production on IP 3 , which is generated through activity of phospholipase C (PLC) [17]. Figure 2A 2+ ] i rises, whereas its inactive analog U73343 (10 mM) had no effect ( Figure 3A-3D). These results suggest that the action of H 2 O 2 mediated through IP 3 , which binds to IP 3 receptors to release Ca 2+ from intracellular Ca 2+ stores. This was confirmed by experiments that measures IP 3 production ( Figure 4). Treatment of cells with H 2 O 2 caused a H 2 O 2 concentration-dependent increase in IP 3 levels in both types of endothelial cells (Figure 4).  Figure 5F) but not in aortic ECs ( Figure 5E).   Figure 6A and 6C). For MAECs, treatment with a lower concentration (500 mM, 26-30 min) was enough to abolish the [Ca 2+ ] i responses to ATP ( Figure 6B and 6D).
To further confirm the findings, Mag-fluo4/AM, a dye that stains Ca 2+ in intracellular Ca 2+ stores, was used to directly measure the store Ca 2+ content. As shown in Figure 6E-6F, treatment with 500 mM H 2 O 2 for 26-30 min caused a marked reduction of store Ca 2+ content by 3366% (n = 3) in MAECs but had no significant effect in aortic ECs. These data suggest that MAECs were more sensitive to H 2 O 2 treatment than aortic ECs with regard to their responses in Ca 2+ store depletion and ATPinduced [Ca 2+ ] i rises. The controls in Figure 6E and 6F were time controls, in which the cells went through 30 min incubation in the absence of H 2 O 2 . In time control, Mag-fluo4 fluorescence only decreased by 866% (n = 3) in MAECs and by 367% (n = 3) in aortic ECs. The small reduction in Mag-fluo4 fluorescence in the control experiments could be due to light-sensitive quenching of Mag-fluo4 as described elsewhere [18].

[Ca 2+ ] i responses to ATP in the absence of H 2 O 2
We also compared ATP-induced Ca 2+ store release in aortic ECs and MAECs in the absence of H 2 O 2 pretreatment. Cells bathed in a nominal Ca 2+ -free solution were challenged with different concentrations of ATP. In both cell types, ATP evoked [Ca 2+ ] i rises in a concentration dependent manner ( Figure 7). Furthermore, the [Ca 2+ ] i response in MAECs was more sensitive to ATP than that in aortic ECs ( Figure 7).

HX-XO-induced [Ca 2+ ] i rises were caused by superoxide anion and hydrogen peroxide
Effect of HX-XO on [Ca 2+ ] i was also studied. HX-XO reacts to yield superoxide anions, which may spontaneously or enzymatically dismutate into H 2 O 2 [4]. Application of HX-XO (200 mM and 20 mU/ml, respectively) evoked rapid [Ca 2+ ] i rises in both types of endothelial cells. Pre-incubation of the cells for 20 min

Discussion
[Ca 2+ ] i change is an important early signal for ROS-induced endothelial function and dysfunction. However, only a few studies have investigated ROS-induced Ca 2+ signaling in the endothelial cells derived from small-sized arteries [8,19] [20]. Indeed, we found that H 2 O 2 treatment could enhance Ca 2+ entry when the bath solution contained Ca 2+ .
There are conflicts in reports as to how ROS treatment would affect the [Ca 2+ ] i responses to subsequent agonist challenge in endothelial cells [7,9,12,15]. In some studies, H 2 O 2 and superoxide anions were found to reduce the agonist-  induced [Ca 2+ ] i rises [9,15]. In other studies, ROS treatment was found to enhance [9,12] or have no effect on the agonistinduced [Ca 2+ ] i responses [7]. In the present study, we found that H 2  ] i responses to ATP. This type of differential sensitivity/response of store Ca 2+ release to ROS treatment could explain some data conflicts in the literature. For example, Volk et al., reported that, in rat liver artery endothelial cells, ROS treatment had no effect on the [Ca 2+ ] i responses to subsequent ATP or histamine challenge [7]. But they used a relatively low concentration of ROS [7]. It is possible that such a low concentration of ROS might not be sufficient to cause marked store Ca 2+ depletion. As a result, no change in [Ca 2+ ] i responses to agonists would be expected. What could be the underlying cellular mechanism for the higher sensitivity of [Ca 2+ ] i responses to H 2 O 2 in MAECs than in aortic ECs? H 2 O 2 -induced IP 3 production was similar in MAECs and aortic ECs, therefore IP 3 production was not the reason. Alternatively, this could be due to more abundant IP 3 receptor expression and/or a higher IP 3 receptor sensitivity to IP 3 in MAECs than in aortic ECs. If this is true, [Ca 2+ ] i responses to other agonists is also expected to be higher in MAECs than in aortic ECs. Indeed, we found that similar high sensitivity of intracellular store Ca 2+ release to ATP in MAECs than in aortic ECs (Figure 7). Therefore, we speculate that MAECs may express more IP 3 receptors and/or the sensitivity of IP 3 receptors to IP 3 may be higher in MAECs than in aortic ECs.
The higher sensitivity of [Ca 2+ ] i responses to H 2 O 2 in the endothelial cell of small-sized arteries could have physiological and/or pathological implication. At physiological concentration, H 2 O 2 is a vasodilator and it causes endothelium-dependent and endothelium-independent vascular dilation [3,23,24]. The effect of H 2 O 2 as a vascular dilator is often found in small-sized arteries and arterioles [3,25]. In contrast, in large-sized arteries nitric acid is a more important vascular dilator [26]  accumulation may lead to endothelial cell apoptosis and cell death [4]. Therefore, it is possible that endothelial cells in small-sized arteries or arterioles might be more vulnerable to ROS-induced cell damage.
H 2 O 2 can be converted to hydroxyl radical in the presence of Fe 2+ [4]. However, in the present study the effect of H 2 O 2 on [Ca 2+ ] i rises in endothelial cells could not be attributed to hydroxyl radical, because the H 2 O 2 effect was not affected by DMSO, which is an efficient hydroxyl radical scavenger [21]. In contrast, H 2 O 2 effect was abolished by catalase, which converts H 2 O 2 to O 2 and H 2 O, suggesting a direct action of H 2 O 2 . We also investigated the effect of HX-XO on [Ca 2+ ] i in mouse aortic ECs and MAECs. HX-XO is one of most widely used methods to generate superoxide anions, which may in turn dismutate into H 2 O 2 spontaneously or enzymatically [4]. We found that the HX-XOinduced [Ca 2+ ] i rises could be attributed to involvement of superoxide anions and H 2 O 2 but not hydroxyl radicals in both types of endothelial cells, because the response was reduced by SOD and catalase but not by DMSO. There were relatively more H 2 O 2 contribution in HX-XO-induced [Ca 2+ ] i rises in endothelial cells of small-sized arteries (MAECs) than in those of large-sized arteries (aortic ECs). Previously, different reports have claimed different ROS, including H 2 O 2 [5,7,10], hydroxyl radical [10], and/or superoxide anions [5,10], to be the contributing factors that were involved in HX-XO provoked-[Ca 2+ ] i rises in endothelial cells. The discrepancy in results could be due to a variety of factors including endothelial cell sources and/or culture conditions.  In conclusion, we found both Ca 2+ entry and store Ca 2+

Ethics statement
We followed Guide for Animal Care and Use of Laboratory Animals published by the US National Institute of Health. The protocols for animal experiments were approved by Animal Experimentation Ethics Committee, The Chinese University of Hong Kong (approval number# 09/060/MIS). Primary cultured aortic endothelial cells (aortic ECs) and mesenteric artery endothelial cells (MAECs) were dissociated from mouse aorta and mesenteric arteries of the first to tertiary branches (internal diameter = 60-200 mm), respectively, using the methods described elsewhere [27]. Aortic ECs and MAECs were cultured in endothelial cell growth medium supplemented with 1% bovine brain extract.

[Ca 2+ ] i Measurement
Cells were prepared and loaded with a membrane permeant fluorescence dye Fluo4/AM (Molecular Probes, Inc., NJ) for observing their [Ca 2+ ] i responses to H 2 O 2 or HX-XO or ATP. Briefly, the cells were seeded on circular glass discs at 37uC overnight supplemented with the culture medium. For the fluorescence dye loading, cells were incubated for 1 hr in dark at room temperature with 10 mM Fluo4/AM and 0.02% Pluronic acid F-127 in normal physiological saline solution (N-PSS), which contained in mM: 1 CaCl 2 , 140 NaCl, 1 KCl, 1 MgCl 2 , 10 glucose, and 5 Hepes at pH 7.4. The circular discs containing the endothelial cells were then pinned in a specially designed chamber. The chamber was placed on the stage of an inverted microscope (Nikon Diaphot 200). During experiments, cells were bathed in N-PSS or 0.5Ca 2+ -PSS or 0Ca 2+ -PSS. The composition of 0.5Ca 2+ -PSS and 0Ca 2+ -PSS was similar to N-PSS except for Ca 2+ concentration (0.5 mM CaCl 2 for 0.5Ca 2+ -PSS, and nominal Ca 2+ -free for 0Ca 2+ -PSS). All agents were applied directly to the bath along the side of the chamber. Solutions were then mixed by pipetting gently up and down for a few times. Experiments were performed at room temperature. Fluorescence signals were recorded by MRC-1000 Laser Scanning Confocal Imaging System with MRC-1000 software (Bio-Rad) with the excitation wavelength of 488 nm and a 515 nm-long pass emission filter. The Ca 2+ responses were displayed as the ratio of fluorescence relative to the intensity before H 2 O 2 or ATP or HX-XO (F1/F0). Due to variation in [Ca 2+ ] i responses between different batches of cells, each series of experiments had its own control.

Measuring Ca 2+ Content in Intracellular Ca 2+ Stores
Cells were loaded with fluorescence dye Mag-fluo4/AM (Molecular Probes, Inc., NJ) for observing the Ca 2+ level in intracellular Ca 2+ stores. Briefly, cells were seeded on circular glass plates at 37uC overnight supplemented with the culture medium. As for the fluorescence dye loading, cells were incubated with 5 mM Mag-fluo4/AM in dark at 37uC for 45 min, and 0.02% Pluronic acid F-127 in N-PSS. Cells were then washed with the indicator-free N-PSS and incubated at 37uC for 45 min to unload the Mag-fluo4 from cytoplasm. The circular discs containing the endothelial cells were then pinned down in a specially designed chamber. The chamber was placed on the stage of an inverted microscope (Nikon Diaphot 200). Mag-fluo4 fluorescence was recorded by MRC-1000 Laser Scanning Confocal Imaging System with MRC-1000 software (Bio-Rad) with the excitation wavelength of 488 nm and a 515 nm-long pass emission filter. The cells were then treated with or without H 2 O 2 for 30 minutes. Because Mag-fluo4 fluorescence was reported to be light-sensitive and could be quenched by light exposure, laser emission to samples was cut off during the period of H 2 O 2 treatment. Fluorescence signals were then collected before and after 30minute H 2 O 2 treatment. The change in store Ca 2+ content is displayed as Mag-fluo4 intensity change in percentage.

IP 3 measurement
The amount of IP 3 was measured using HitHunter TM IP 3 Assay Fluorescence Polarization Detection-Green Kits (DiscoveRx Tech, Fremont, CA, USA), a reliable and convenient methodology based on competitive binding between an IP 3 fluorescence tracer and unlabeled IP 3 from the cell lysates or standards. Free IP 3 competes at the IP 3 binding protein and allows the IP 3 tracer to rotate freely upon excitation with plane polarized light. The polarized signal is inversely proportional to the amount of the free unlabelled IP 3 . Thus, polarization signal is decreased when the concentration of free IP 3 is increased [22]. Briefly, aortic ECs and MAECs were treated with different concentrations of H 2 O 2 (500 mM, 2 mM, 5 mM) for 5 min in black 96-well plates. The cellular reactions were terminated by placing cells on ice followed by addition of 0.2 N perchloric acid to lyse the cells. The plate was then shaken at 650 rpm for 5 min. The IP 3 tracer was subsequently added to each well, followed by IP 3 binding protein. The polarized fluorescence from the IP 3 tracer (fluorescein) was read using a Wallac EnVision TM Microplate Reader (Perkin Elmer, Wallac, EnVision, Finland) with a polarization mirror, a 485 nm excitation filter and a 530 nm emission filter. IP 3 concentration was calculated from the IP 3 standard curve and expressed as pmole/1610 6 cells.

Data Analysis
Data Analysis was performed with Software Confocal Assistant and Metafluor. All representative traces were plotted by using Prism 3.0 (GraphPad, San Diego, CA, USA). Summarized data were expressed as the mean6SEM and analyzed with two-tailed Student's t test at a p,0.05 level of significance.