Critical Role of Phospholipase Cγ1 in the Generation of H2O2-evoked [Ca2+]i Oscillations in Cultured Rat Cortical Astrocytes*

Reactive oxygen species, such as the superoxide anion, H2O2, and the hydroxyl radical, have been considered as cytotoxic by-products of cellular metabolism. However, recent studies have provided evidence that H2O2 serves as a signaling molecule modulating various physiological functions. Here we investigated the effect of H2O2 on the regulation of intracellular Ca2+ signaling in rat cortical astrocytes. H2O2 triggered the generation of oscillations of intracellular Ca2+ concentration ([Ca2+]i) in a concentration-dependent manner over the range 10–100μm. The H2O2-induced [Ca2+]i oscillations persisted in the absence of extracellular Ca2+ and were prevented by depletion of intracellular Ca2+ stores with thapsigargin. The H2O2-induced [Ca2+]i oscillations were not inhibited by pretreatment with ryanodine but were prevented by 2-aminoethoxydiphenyl borate and caffeine, known antagonists of inositol 1,4,5-trisphosphate receptors. H2O2 activated phospholipase C (PLC) γ1 in a dose-dependent manner, and U73122, an inhibitor of PLC, completely abolished the H2O2-induced [Ca2+]i oscillations. In addition, RNA interference against PLCγ1 and the expression of the inositol 1,4,5-trisphosphate-sequestering “sponge” prevented the generation of [Ca2+]i oscillations. H2O2-induced [Ca2+]i oscillations and PLCγ1 phosphorylation were inhibited by pretreatment with dithiothreitol, a sulfhydryl-reducing agent. Finally, epidermal growth factor induced H2O2 production, PLCγ1 activation, and [Ca2+]i increases, which were attenuated by N-acetylcysteine and diphenyleneiodonium and by the overexpression of peroxiredoxin type II. Therefore, we conclude that low concentrations of exogenously applied H2O2 generate [Ca2+]i oscillations by activating PLCγ1 through sulfhydryl oxidation-dependent mechanisms. Furthermore, we show that this mechanism underlies the modulatory effect of endogenously produced H2O2 on epidermal growth factor-induced Ca2+ signaling in rat cortical astrocytes.

H 2 O 2 is a member of the reactive oxygen species (ROS), 6 which cause oxidative damage to cellular components such as lipids, nucleic acids, and proteins. Therefore, H 2 O 2 has generally been considered to be cytotoxic and hazardous to living organisms. However, a growing body of evidence suggests that H 2 O 2 serves as an intracellular signaling molecule modulating various physiological functions (1). Cells possess mechanisms that can rapidly synthesize and destroy H 2 O 2 in response to receptor stimulation. For example, stimulation of membrane receptors of various growth factors, such as transforming growth factor-␤1, platelet-derived growth factor, and epidermal growth factor (EGF) triggers the rapid and transient production of H 2 O 2 (2)(3)(4)(5). H 2 O 2 generated in response to receptor stimulation has been shown to play an important role in regulating various normal cell functions, such as cell proliferation, platelet aggregation, and vasodilation (6 -8). In addition to this, exogenous addition of H 2 O 2 at low concentrations affects the functions of various ion channels and other proteins involved in signal transduction (8 -10). Therefore, H 2 O 2 fulfills the prerequisites for being considered as a genuine intracellular messenger.
Recently, a great deal of attention has focused on the sensitivity of the mechanisms responsible for Ca 2ϩ mobilization in response to changes in the cellular redox state. Ca 2ϩ plays a pivotal role in the regulation of a diverse range of cellular functions, such as muscle contraction, secretion, synaptic plasticity, cell proliferation, and cell death (11). Many hormones and neurotransmitters increase intracellular Ca 2ϩ concentration ([Ca 2ϩ ] i ) by mobilizing Ca 2ϩ from intracellular stores and by inducing an influx from the extracellular space (12,13). H 2 O 2 has been shown to enhance the activity of L-type Ca 2ϩ channels (10). Peroxide can also stimulate the mobilization of Ca 2ϩ in many cell types by modifying Ca 2ϩ release channels, such as TRPM2 (14), ryanodine receptors (15), and inositol 1,4,5-trisphosphate (IP 3 )-dependent Ca 2ϩ channels (16). In addition, H 2 O 2 can modify the activity of Ca 2ϩ pumps involved in Ca 2ϩ homeostasis, such as the sarcoendoplasmic reticulum Ca 2ϩ -ATPase (SERCA) (17) and plasma membrane Ca 2ϩ -ATPase (17). Furthermore, enzymes involved in Ca 2ϩ signaling pathways, such as phospholipase C (PLC) ␥1 (18) and phospholipase D (19) are also targets. However, most of the previous studies employed high concentrations of H 2 O 2 , and it is questionable whether such diverse actions of H 2 O 2 on calcium signaling also occur under normal physiological conditions. Therefore, it will be of great value to identify the target molecules modulated by physiologically relevant concentrations of H 2 O 2 .
Astrocytes, the major glial cell type in the mammalian brain, participate in a variety of important functions in the central nervous system. As in virtually all other cell types, astrocytes also use Ca 2ϩ signaling to mediate a large spectrum of physiological responses. Elevation of [Ca 2ϩ ] i in response to stimulation of various receptors causes the release of neurotransmitters, such as glutamate and ATP, and plays an important role in the exchange of information with neurons and the regulation of local blood flow (20,21). In contrast to Ca 2ϩ signaling, much less attention has been given to redox signaling in astrocytes. Because ROS are involved in the pathogenesis of neurodegenerative diseases and astrocytes have been shown to possess high antioxidant activities, many studies have focused on the protective roles of astrocytes against oxidative stress-induced neuronal cell death (22)(23)(24). However, despite the lack of information about the physiological roles of ROS in astrocytes, NADPH oxidase was shown to be involved in the generation of H 2 O 2 and cell survival in this cell type (25). Given the widespread involvement of H 2 O 2 in modulating Ca 2ϩ signaling cascades, it is tempting to speculate that astrocytes may also use redox signaling to modify their Ca 2ϩ signaling.
Therefore, in the present study, we sought to investigate the roles of H 2 O 2 in Ca 2ϩ signaling in cultured rat astrocytes. Our results indicate that a low, physiologically relevant concentration of H 2 O 2 (30 M) induces [Ca 2ϩ ] i oscillations in a PLC␥1-and IP 3 -dependent manner. In addition, H 2 O 2 produced endogenously by EGF receptor stimulation is involved in the modulation of Ca 2ϩ signaling in rat astrocytes.
Cell Cultures-Primary cultures of cortical astrocytes were prepared from neonatal (12-24 h) Wistar rats. Briefly, cortices were dissected, and the tissues were minced and mechanically dissociated. Then the isolated cells were plated on 60-mm culture dishes and maintained at 37°C in a humidified 5% CO 2 and 95% air for 2-3 weeks. For [Ca 2ϩ ] i measurements, cells were cultured on 0.01% poly-L-lysine-coated cover glasses in 60-mm dishes for 7-10 days. The culture medium consisted of MEM supplemented with 2 mM glutamine, 25 mM glucose, 100 g/ml penicillin, 25 ng/ml streptomycin, and 10% fetal bovine serum. The culture medium was replaced every 3 days. Cells were serum-starved for 2 days before each experiment.
Expression of IP 3 Sponge and Peroxiredoxin Type II (Prx II)-Astrocytes were transiently transfected with a green fluorescent protein (GFP)-tagged high affinity (R441Q) or low affinity (K508A) IP 3 -sequestering sponge (26), or were cotransfected with Prx II (1 g/ml; kindly provided by Professor S. W. Kang, Ewha Womans University, Seoul, Korea) and eGFP-N1 (1.2 g/ml; Clontech) using Lipofectamine 2000 reagent (Invitrogen), according to the manufacturer's instructions. Cells were incubated for 48 h at 37°C, in a 5% CO 2 atmosphere with saturated humidity to allow expression of the construct before the experiment. The expression of each protein was confirmed by GFP fluorescence.
Western Blot Analysis-Astrocytes transfected with or without siRNA-PLC␥1 and eGFP-N1 were stimulated with H 2 O 2 or EGF for the indicated times in the physiological salt solutions (PSSs) containing 140 mM NaCl, 5 mM KCl, 1 mM MgCl 2 , 1 mM CaCl 2 , 10 mM HEPES, and 10 mM glucose (pH 7.4). Cells were then lysed at 4°C in lysis buffer (150 mM NaCl, 1% Triton X-100, 50 mM Tris, 10 mM NaF, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 1 g/ml aprotinin, 1 mM leupeptin, and 1 mM sodium orthovanadate) and centrifuged at 12,000 rpm for 10 min at 4°C. The samples were subjected to SDS-PAGE and subsequently transferred to nitrocellulose membranes. The membranes were incubated with specific antibodies against PLC␥1, phosphospecific tyrosine 783, PLC␤1, and actin, and the proteins were detected by ECL (Amersham Biosciences). The intensity of bands was quantified using Meta-Morph Analysis System (Universal Imaging Co., Downingtown, PA).
[Ca 2ϩ ] i Measurements-For [Ca 2ϩ ] i measurements, attached cells were loaded with fura-2 by incubation with 3.5 M fura-2-acetoxymethyl ester in PSS equilibrated with 100% O 2 for 40 min at room temperature. The cells were then washed twice and rested for at least 20 min before use. The fura-2-loaded cells were mounted on the stage of an inverted microscope (Nikon, Tokyo, Japan) for imaging. The cells were superfused at a constant perfusion rate with the PSS. In Ca 2ϩ -free solutions, CaCl 2 was omitted, and 1 mM EGTA was added. The excitation wavelength was alternated between 340 and 380 nm, and the emission fluorescence was monitored at 510 nm with a CCD camera using MetaFluor system (Universal Imaging Co., Downingtown, PA). Fluorescence images were obtained at 4-s intervals. Background fluorescence was subtracted from the raw signals at each excitation wavelength, and the values of [Ca 2ϩ ] i were calculated using the equation described previously (28).
ROS Imaging-ROS levels were measured using the fluorescence probe DCF. In brief, cells were incubated for 5 min in the presence of 5g/ml DCF and washed in Hanks' balanced salt solution. DCF fluorescence was measured using a confocal laser-scanning microscope (Leica, Buffalo, NY) with an excitation wavelength at 488 nm and an emission at 525 nm. To avoid photo-oxidation of DCF, the fluorescence images were collected using a single rapid scan, and identical settings were used for all samples.
Data Analysis-The results are presented as mean Ϯ S.E. Statistical analysis was performed by unpaired Student's t test. p values lower than 0.05 were considered to be statistically significant. In general, as shown in Fig. 1 (Fig. 2B). To examine whether the thapsigargin-sensitive intracellular Ca 2ϩ store was releasable by IP 3 receptors or ryanodine receptors, cells were exposed to 75 M 2-APB, 20 mM caffeine (IP 3 receptor antagonists), or 100 M ryanodine (a ryanodine receptor antagonist), prior to the addition of 30 M H 2 O 2 . As shown in Fig. 2C   phorylated following H 2 O 2 stimulation of cultured rat astrocytes. PLC␥1 possesses three tyrosine residues, Tyr-771, Tyr-783, and Tyr-1254. Among them, Tyr-783 is known to be essential for IP 3 formation (31). Therefore, a phosphospecific tyrosine 783 antibody was used to detect the H 2 O 2 -induced phosphorylation of PLC␥1. As shown in Fig. 3, A and B, exposure of the astrocytes to various concentrations of H 2 O 2 for 10 min induced PLC␥1 phosphorylation on tyrosine residue 783 in a dose-dependent manner (n ϭ 6).

H 2 O 2 Mobilizes Ca 2ϩ in Cultured Rat Cortical Astrocytes in a Con
To clarify further the involvement of PLC␥1 in the generation of [Ca 2ϩ ] i oscillations, we used the PLC inhibitor U73122 and as control its inactive analogue U73343. As shown in Fig. 3 The involvement of PLC␥1 was further investigated by using RNA interference (RNAi). As shown in Fig. 4, A and B, transfection of siRNA-PLC␥1 suppressed the PLC␥1 expression level, although that of PLC␤1 expression was not affected (n ϭ 5). To examine the functional consequences of depletion of PLC␥1 by RNAi, cells were transfected with pSUPER (empty vector) or siRNA-PLC␥1 prior to measurement of To confirm whether the transfection of siRNA-PLC␥1 had any effect on the Ca 2ϩ response elicited by a PLC␤-activating agonist, we stimulated the cells with histamine. As shown in Fig. 4, E and F, the cotransfection of GFP and siRNA-PLC␥1 did not prevent the Ca 2ϩ responses to histamine (n ϭ 6). These results indicate that PLC␥1 is necessary for the generation of [Ca 2ϩ ] i oscillations in response to H 2 O 2 .
In addition, we also observed that expression of the IP 3 sponge completely abrogated [Ca 2ϩ ] i oscillations in response to H 2 O 2 (n ϭ 8; see   Fig. 6, A and B, DTT (n ϭ 16; Fig. 6C). These data suggest that   Fig. 7, A and B, EGF at a concentration of 10 ng/ml induced an increase in DCF fluorescence intensity that was prevented by 5 mM NAC, indicating that EGF increased an accumulation of ROS (n ϭ 5). The DCF fluorescence intensity was also decreased by 10 M DPI (an inhibitor of NADPH oxidase), suggesting that NADPH oxidase, at least in part, participated in the EGF-triggered generation of ROS (n ϭ 5).
EGF induced a rapid transient peak increase in [Ca 2ϩ ] i that subsequently declined (n ϭ 15; Fig. 8A). However, pretreatment with 5 mM NAC or 10 M DPI attenuated the EGF-induced [Ca 2ϩ ] i increases, and removal of NAC and DPI in the continued presence of EGF increased [Ca 2ϩ ] i again (n ϭ 16 -20; Fig. 8, B and C). The effect of NAC and DPI on the EGF-induced activation of PLC␥1 was also investigated. As shown in Fig. 8, D and E, the immediate strong activation of PLC␥1 followed by a sustained weak activation was observed following EGF stimulation, but in the presence of 5 mM NAC or 10 M DPI the activation of PLC␥1 was greatly reduced (n ϭ 3).
Because Prx II is a cellular peroxidase that eliminates endogenous H 2 O 2 produced in response to growth factors such as EGF (33), we examined whether the overexpression of Prx II also attenuated EGFinduced [Ca 2ϩ ] i oscillations. As shown in Fig. 8, F and G, overexpression of Prx II decreased the amplitude of EGF-induced [Ca 2ϩ ] i increase by about 57% (n ϭ 6). These data suggested that H 2 O 2 is generated by the activation of NADPH oxidase and is subsequently involved in the activation of PLC and the elevation of [Ca 2ϩ ] i during EGF stimulation in cultured rat astrocytes.

DISCUSSION
In this study, we report that exogenous addition of low concentrations of H 2 O 2 triggers [Ca 2ϩ ] i oscillations through the activation of PLC␥1 in cultured rat cortical astrocytes. H 2 O 2 -mediated elevation of cytosolic Ca 2ϩ levels has been shown previously in various cell types (32)(33)(34)(35)(36). However, in many cases, [Ca 2ϩ ] i increases were induced by relatively high concentrations of H 2 O 2 , which are generally considered cytotoxic. The physiologically relevant concentration range of H 2 O 2 , which causes an acceleration of cellular functions in a variety of cell types, is considered 1-100 M, although it depends on cell type (1,37 In most nonexcitable cells, such as astrocytes, both Ca 2ϩ release from intracellular Ca 2ϩ stores and Ca 2ϩ influx through Ca 2ϩ channels on the plasma membrane are necessary for the generation and maintenance of [Ca 2ϩ ] i oscillations (40). In the present study, we demonstrate that H 2 O 2 -induced [Ca 2ϩ ] i oscillations were sustained in the absence of extracellular Ca 2ϩ , indicating that intracellular Ca 2ϩ stores were primarily responsible for the generation of [Ca 2ϩ ] i oscillations. The two main intracellular organelles containing large amounts of Ca 2ϩ are the endoplasmic reticulum and mitochondria (41). Previously, both of these Ca 2ϩ stores were shown to be involved in H 2 O 2 -induced [Ca 2ϩ ] i increases (17). However, our data showed that depletion of intracellular Ca 2ϩ stores with thapsigargin completely prevented the generation of H 2 O 2 -evoked [Ca 2ϩ ] i oscillations, suggesting that the thapsigargin-sensitive endoplasmic reticulum Ca 2ϩ store was the source of [Ca 2ϩ ] i oscillations.
H 2 O 2 was also reported to be involved in the mobilization of Ca 2ϩ by activating intracellular Ca 2ϩ channels, such as ryanodine receptors and IP 3 receptors (15,16). The effect of ROS on ryanodine receptors has been well established. Sulfhydryl oxidation of ryanodine receptors has been reported to activate the channels (42,43). However, in the present study, a high concentration of ryanodine (100 M), which blocked the   . EGF produces ROS by the activation of NADPH oxidase in cultured rat cortical astrocytes. A, astrocytes were loaded with DCF for 5 min (a-d). Cells were treated with 5 mM NAC (c) or 10 M DPI (d) for 2 min followed by an addition of 10 ng/ml EGF (b-d) for 2 min. The fluorescence of DCF was subsequently visualized by confocal laser scanning microscopy. B, the DCF fluorescence was quantified, and the relative intensities were calculated by setting the fluorescence intensity of control cells to 100% (n ϭ 5). Results are means Ϯ S.E. targets; it blocks IP 3 -sensitive Ca 2ϩ channels, SERCA activity, and capacitative Ca 2ϩ entry channels (44,45). However, the inhibitory effect of 2-APB on [Ca 2ϩ ] i oscillations was unlikely to be due to the inhibition of SERCA, because the concentration of 2-APB we used in this study (75 M) was lower than the half-maximal inhibitory concentration for SERCA (91 M) (44). In addition, 75 M 2-APB did not show any evidence of [Ca 2ϩ ] i increase when applied to itself. This is in contrast to 1 M thapsigargin, a specific SERCA inhibitor, which induced a rapid increase in [Ca 2ϩ ] i as shown in Fig. 2B. Inhibition of SERCA has been shown to be associated with an increase in [Ca 2ϩ ] i in most cell types. Furthermore, it is unlikely that the effect of 2-APB on the [Ca 2ϩ ] i oscillations was because of inhibition of capacitative Ca 2ϩ entry, because as shown for the experiments performed in the absence of extracellular Ca 2ϩ , Ca 2ϩ entry is not required for the oscillations.
Caffeine also has several cellular targets. It can stimulate ryanodine receptors, inhibit both cAMP degradation and PLC activation, and prevent IP 3 -sensitive Ca 2ϩ channel opening. However, the only feature that caffeine and 2-APB share is their ability to antagonize IP 3 -mediated Ca 2ϩ release. Therefore, although neither 2-APB nor caffeine are solely selective for IP 3 -sensitive Ca 2ϩ channels, when used judiciously these pharmacological agents can be used to reveal the specific involvement of IP 3 signaling. The results obtained using 2-APB and caffeine support the hypothesis that H 2 O 2 induced [Ca 2ϩ ] i oscillations through activation of IP 3 -sensitive Ca 2ϩ channels.
H 2 O 2 may activate signaling components responsible for IP 3 production. In some cell types, it has been reported that H 2 O 2 induces an activation of PLC␥1 (29,30). PLC␥1 is known to be recruited to the plasma membrane following activation of receptor tyrosine kinases and activated by a mechanism that relies on tyrosine phosphorylation (46). The phosphorylated PLC␥1 catalyzes the hydrolysis of phosphatidylinositol 4,5-bisphosphate to produce diacylglycerol and IP 3 , leading to the activation of protein kinase C and increases in [Ca 2ϩ ] i , respectively (47). Recently, PLC␥1 has been reported to be tyrosine-phosphorylated following H 2 O 2 treatment and to protect cells from oxidant injury (18  Note that pretreatment with NAC or DPI suppressed Ca 2ϩ responses to EGF, and the removal of NAC and DPI permitted the Ca 2ϩ responses. Results are representative of 15-20 cells in three or four independent experiments. D, the effect of 5 mM NAC or 10 M DPI on 10 ng/ml EGF-induced PLC␥1 phosphorylation is shown. Cells were incubated with or without NAC or DPI for 2 min, and then 10 ng/ml EGF was added for the indicated times. Cells were then lysed, and the lysates were subjected to immunoblot analysis with antibodies to phosphotyrosine (PY783) or PLC␥1. E, quantitation of results in D. Ratio (PY783/PLC␥1) was calculated (n ϭ 3). F, cells were cotransfected with GFP and Prx II, which eliminates H 2 O 2 , and [Ca 2ϩ ] i was measured. Note that GFP-and Prx II-transfected cell (designated as a) exhibited reduced Ca 2ϩ response to EGF. G, the amplitude of [Ca 2ϩ ] i increases (nM) in cells transfected with or without GFP and Prx II was measured during EGF stimulation in six independent experiments. Results are presented as mean Ϯ S.E. * indicates the difference of the amplitudes of [Ca 2ϩ ] i increases between the Prx II-transfected and nontransfected groups ( p Ͻ 0.05).  (46,49,50). In addition, Hu et al. (16,51) showed that NADPH oxidase-derived H 2 O 2 increased the sensitivity of intracellular Ca 2ϩ stores to IP 3 and played a critical role in generating [Ca 2ϩ ] i oscillations in human endothelial cells stimulated by histamine. Taken together, our study and those of Hu et al. (16,51) imply that both PLC␥1 activation and increased sensitivity of IP 3 receptors may contribute to the generation of [Ca 2ϩ ] i oscillations.
By having observed the stimulatory effect of exogenous H 2 O 2 on intracellular Ca 2ϩ signaling, we sought to determine whether H 2 O 2 was produced by receptor stimulation and if endogenously generated H 2 O 2 played a modulatory role in Ca 2ϩ signaling in rat astrocytes. Stimulation of EGF receptors has been shown previously to induce both H 2 O 2 production and [Ca 2ϩ ] i increases in fibroblasts (32). The EGF receptor belongs to a family of transmembrane receptors with intrinsic tyrosine kinase activity (52). The production of intracellular H 2 O 2 in response to EGF was shown to require the activation of the Rac-NADPH oxidase signaling, whereas activation of PLC␥1 was regarded to be critical for [Ca 2ϩ ] i increases (53)(54)(55). Although Rac was suggested to play a role in the EGF-induced Ca 2ϩ signaling (32,56), the role of H 2 O 2 in PLC␥1 activation during EGF stimulation has not been elucidated.
In this study, we showed that EGF receptor stimulation induced ROS production, PLC␥1 activation, and [Ca 2ϩ ] i elevation, which were all attenuated by the pretreatment with NAC, an ROS scavenger, or DPI, an NADPH oxidase inhibitor. These results indicated that ROS produced via NADPH oxidase during EGF stimulation played a critical role in the enhancement and maintenance of PLC␥1 and Ca 2ϩ responses in rat cortical astrocytes. As far as we know, this is the first report to show the involvement of NADPH oxidase in EGF-mediated ROS generation and the regulatory role of endogenously produced ROS in PLC␥1-activated Ca 2ϩ signaling in astrocytes. Previous studies have revealed that the predominant member of ROS produced by EGF stimulation was H 2 O 2 , which played a key role in the EGF-induced protein tyrosine phosphorylation and [Ca 2ϩ ] i increases (4,32). Furthermore, we found that overexpression of Prx II, which probably plays an important role in eliminating H 2 O 2 generated in the cytosol, reduced the amplitude of [Ca 2ϩ ] i increase evoked by EGF. Therefore, it is likely that the major component of ROS, which is produced by EGF stimulation and responsible for PLC␥1 activation and [Ca 2ϩ ] i elevation in our system, is H 2 O 2 .
Considering that the activation of EGF receptors stimulates the proliferation and differentiation of astrocytes (57,58), the role of H 2 O 2 in the regulation of Ca 2ϩ signaling may be of physiological importance in this cell type.
We therefore conclude that physiologically relevant, low concentrations of H 2 O 2 trigger the generation of [Ca 2ϩ ] i oscillations by activating PLC␥1 through sulfhydryl oxidation-dependent mechanisms in cultured rat cortical astrocytes. Given that H 2 O 2 is a small and diffusible molecule that is produced endogenously via NADPH oxidase during EGF receptor stimulation and is involved in the enhancement of Ca 2ϩ signaling, H 2 O 2 may be of physiological importance in regulating various cellular functions such as cell proliferation.